OPTION VALUES: Model directory: ./IOE2+0.06/ Training data file: train.tagged Testing data file: test.tagged Unlabeled data file: data.untagged Model file: model.txt Training log file (this one): trainlog.txt Label representation: IOE2 Second-order Markov CRFs Number of labels: 24 Number of training sequences: 443 (one data partition) Number of testing sequences: 22 (one data partition) Number of unlabeled sequences: 0 Number of context predicates: 687508 Number of features: 1466312 Feature rare threshold: 1 Context predicate rare threshold: 1 Using multiple rare thresholds for features: 0 Highlight feature: 0 Number of training iterations: 200 Initial lambda value: 0.0600 Sigma square (for smoothing): 100.0000 Epsilon for L-BFGS convergence: 0.000100 Number of approximated hessian matrixes: 7 Start to train ... Iteration: 1 Log-likelihood = -4545069.073377 Norm (log-likelihood gradient vector) = 518809.144219 Norm (lambda vector) = 72.654822 Log-likelihood and gradient computational time: 319 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4818 4498 93.36 97.09 95.19 i-np 13660 15306 13307 86.94 97.42 91.88 e-np 12220 12293 11554 93.99 94.55 94.27 o 6349 5970 5764 96.55 90.79 93.58 e-vp 4768 4518 4377 96.88 91.80 94.27 i-vp 2602 2497 2388 95.63 91.78 93.67 e-adjp 384 140 132 94.29 34.38 50.38 i-pp 52 23 22 95.65 42.31 58.67 e-advp 822 487 448 91.99 54.50 68.45 i-advp 100 20 19 95.00 19.00 31.67 e-sbar 503 237 234 98.73 46.52 63.24 i-adjp 152 22 19 86.36 12.50 21.84 e-prt 126 110 105 95.45 83.33 88.98 i-sbar 12 1 1 100.00 8.33 15.38 i-conjp 24 1 1 100.00 4.17 8.00 e-conjp 16 1 1 100.00 6.25 11.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 7 7 100.00 70.00 82.35 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.04 47.24 59.68 Avg2. 46451 46451 42877 92.31 92.31 92.31 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12303 10300 83.72 84.29 84.00 pp 4633 4818 4483 93.05 96.76 94.87 vp 4768 4518 4235 93.74 88.82 91.21 sbar 503 237 229 96.62 45.53 61.89 adjp 384 141 128 90.78 33.33 48.76 advp 822 487 447 91.79 54.38 68.30 prt 126 110 105 95.45 83.33 88.98 lst 10 7 7 100.00 70.00 82.35 intj 4 0 0 0.00 0.00 0.00 conjp 16 1 1 100.00 6.25 11.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 84.51 56.27 67.56 Avg2. 23486 22622 19935 88.12 84.88 86.47 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 354 seconds Iteration: 2 Log-likelihood = -4037124.537568 Norm (log-likelihood gradient vector) = 495378.495006 Norm (lambda vector) = 72.698472 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4892 3996 81.68 86.25 83.91 i-np 13660 20901 13257 63.43 97.05 76.72 e-np 12220 12130 9910 81.70 81.10 81.40 o 6349 5274 4805 91.11 75.68 82.68 e-vp 4768 2179 1862 85.45 39.05 53.61 i-vp 2602 1069 1054 98.60 40.51 57.42 e-adjp 384 3 3 100.00 0.78 1.55 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 2 2 100.00 0.24 0.49 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 0 0 0.00 0.00 0.00 i-adjp 152 0 0 0.00 0.00 0.00 e-prt 126 1 1 100.00 0.79 1.57 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 40.10 21.07 27.63 Avg2. 46451 46451 34890 75.11 75.11 75.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12137 6642 54.73 54.35 54.54 pp 4633 4892 3967 81.09 85.62 83.30 vp 4768 2179 1775 81.46 37.23 51.10 sbar 503 0 0 0.00 0.00 0.00 adjp 384 3 3 100.00 0.78 1.55 advp 822 2 2 100.00 0.24 0.49 prt 126 1 1 100.00 0.79 1.57 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 51.73 17.90 26.60 Avg2. 23486 19214 12390 64.48 52.75 58.03 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 3 Log-likelihood = -2436151.382173 Norm (log-likelihood gradient vector) = 300353.450782 Norm (lambda vector) = 73.009907 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4853 3996 82.34 86.25 84.25 i-np 13660 22600 13315 58.92 97.47 73.44 e-np 12220 11138 8655 77.71 70.83 74.11 o 6349 5192 4777 92.01 75.24 82.78 e-vp 4768 1804 1533 84.98 32.15 46.65 i-vp 2602 864 851 98.50 32.71 49.11 e-adjp 384 0 0 0.00 0.00 0.00 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 0 0 0.00 0.00 0.00 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 0 0 0.00 0.00 0.00 i-adjp 152 0 0 0.00 0.00 0.00 e-prt 126 0 0 0.00 0.00 0.00 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 24.72 19.73 21.95 Avg2. 46451 46451 33127 71.32 71.32 71.32 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11144 5501 49.36 45.02 47.09 pp 4633 4853 3965 81.70 85.58 83.60 vp 4768 1804 1378 76.39 28.90 41.94 sbar 503 0 0 0.00 0.00 0.00 adjp 384 0 0 0.00 0.00 0.00 advp 822 0 0 0.00 0.00 0.00 prt 126 0 0 0.00 0.00 0.00 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 20.75 15.95 18.03 Avg2. 23486 17801 10844 60.92 46.17 52.53 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 4 Log-likelihood = -1687966.773265 Norm (log-likelihood gradient vector) = 302183.612595 Norm (lambda vector) = 74.428001 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4951 4297 86.79 92.75 89.67 i-np 13660 16150 12904 79.90 94.47 86.57 e-np 12220 13178 11007 83.53 90.07 86.68 o 6349 6081 5591 91.94 88.06 89.96 e-vp 4768 4053 3544 87.44 74.33 80.35 i-vp 2602 2038 1899 93.18 72.98 81.85 e-adjp 384 0 0 0.00 0.00 0.00 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 0 0 0.00 0.00 0.00 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 0 0 0.00 0.00 0.00 i-adjp 152 0 0 0.00 0.00 0.00 e-prt 126 0 0 0.00 0.00 0.00 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 26.14 25.63 25.88 Avg2. 46451 46451 39242 84.48 84.48 84.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 13185 9140 69.32 74.80 71.95 pp 4633 4951 4261 86.06 91.97 88.92 vp 4768 4053 3337 82.33 69.99 75.66 sbar 503 0 0 0.00 0.00 0.00 adjp 384 0 0 0.00 0.00 0.00 advp 822 0 0 0.00 0.00 0.00 prt 126 0 0 0.00 0.00 0.00 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 23.77 23.68 23.72 Avg2. 23486 22189 16738 75.43 71.27 73.29 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 5 Log-likelihood = -1277725.491439 Norm (log-likelihood gradient vector) = 152155.468825 Norm (lambda vector) = 74.673585 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 5098 4434 86.98 95.70 91.13 i-np 13660 14845 12932 87.11 94.67 90.73 e-np 12220 13087 11455 87.53 93.74 90.53 o 6349 5999 5671 94.53 89.32 91.85 e-vp 4768 4715 4111 87.19 86.22 86.70 i-vp 2602 2707 2296 84.82 88.24 86.49 e-adjp 384 0 0 0.00 0.00 0.00 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 0 0 0.00 0.00 0.00 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 0 0 0.00 0.00 0.00 i-adjp 152 0 0 0.00 0.00 0.00 e-prt 126 0 0 0.00 0.00 0.00 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 26.41 27.39 26.89 Avg2. 46451 46451 40899 88.05 88.05 88.05 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 13089 10126 77.36 82.86 80.02 pp 4633 5098 4394 86.19 94.84 90.31 vp 4768 4715 3819 81.00 80.10 80.54 sbar 503 0 0 0.00 0.00 0.00 adjp 384 0 0 0.00 0.00 0.00 advp 822 0 0 0.00 0.00 0.00 prt 126 0 0 0.00 0.00 0.00 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 24.46 25.78 25.10 Avg2. 23486 22902 18339 80.08 78.08 79.07 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 6 Log-likelihood = -1060365.462225 Norm (log-likelihood gradient vector) = 84849.970212 Norm (lambda vector) = 74.906201 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 5129 4461 86.98 96.29 91.40 i-np 13660 14240 12960 91.01 94.88 92.90 e-np 12220 12819 11692 91.21 95.68 93.39 o 6349 5978 5680 95.02 89.46 92.16 e-vp 4768 4946 4224 85.40 88.59 86.97 i-vp 2602 3251 2457 75.58 94.43 83.96 e-adjp 384 0 0 0.00 0.00 0.00 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 85 82 96.47 9.98 18.08 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 3 3 100.00 0.60 1.19 i-adjp 152 0 0 0.00 0.00 0.00 e-prt 126 0 0 0.00 0.00 0.00 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 36.08 28.49 31.84 Avg2. 46451 46451 41559 89.47 89.47 89.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12820 10647 83.05 87.13 85.04 pp 4633 5129 4421 86.20 95.42 90.58 vp 4768 4946 3832 77.48 80.37 78.90 sbar 503 3 3 100.00 0.60 1.19 adjp 384 0 0 0.00 0.00 0.00 advp 822 85 82 96.47 9.98 18.08 prt 126 0 0 0.00 0.00 0.00 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 44.32 27.35 33.83 Avg2. 23486 22983 18985 82.60 80.84 81.71 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 7 Log-likelihood = -907314.562555 Norm (log-likelihood gradient vector) = 69911.963220 Norm (lambda vector) = 75.389473 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4855 4473 92.13 96.55 94.29 i-np 13660 13931 13066 93.79 95.65 94.71 e-np 12220 12497 11824 94.61 96.76 95.68 o 6349 6062 5798 95.65 91.32 93.43 e-vp 4768 5002 4400 87.96 92.28 90.07 i-vp 2602 3354 2518 75.07 96.77 84.55 e-adjp 384 100 97 97.00 25.26 40.08 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 407 338 83.05 41.12 55.00 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 240 232 96.67 46.12 62.45 i-adjp 152 3 3 100.00 1.97 3.87 e-prt 126 0 0 0.00 0.00 0.00 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 45.80 34.19 39.15 Avg2. 46451 46451 42749 92.03 92.03 92.03 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12502 11035 88.27 90.30 89.27 pp 4633 4855 4434 91.33 95.70 93.47 vp 4768 5003 3997 79.89 83.83 81.81 sbar 503 240 228 95.00 45.33 61.37 adjp 384 100 88 88.00 22.92 36.36 advp 822 407 304 74.69 36.98 49.47 prt 126 0 0 0.00 0.00 0.00 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 51.72 37.51 43.48 Avg2. 23486 23107 20086 86.93 85.52 86.22 Current max chunk-based F1: 86.47 (iteration 1) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 8 Log-likelihood = -722616.358943 Norm (log-likelihood gradient vector) = 71611.624836 Norm (lambda vector) = 76.886290 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4819 4491 93.19 96.94 95.03 i-np 13660 13404 12893 96.19 94.39 95.28 e-np 12220 12572 11938 94.96 97.69 96.31 o 6349 6465 6090 94.20 95.92 95.05 e-vp 4768 4890 4510 92.23 94.59 93.39 i-vp 2602 2990 2511 83.98 96.50 89.81 e-adjp 384 262 220 83.97 57.29 68.11 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 679 522 76.88 63.50 69.55 i-advp 100 0 0 0.00 0.00 0.00 e-sbar 503 324 304 93.83 60.44 73.52 i-adjp 152 27 23 85.19 15.13 25.70 e-prt 126 19 19 100.00 15.08 26.21 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 49.73 39.37 43.95 Avg2. 46451 46451 43521 93.69 93.69 93.69 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12580 11241 89.36 91.99 90.65 pp 4633 4819 4451 92.36 96.07 94.18 vp 4768 4892 4247 86.82 89.07 87.93 sbar 503 324 300 92.59 59.64 72.55 adjp 384 262 182 69.47 47.40 56.35 advp 822 679 477 70.25 58.03 63.56 prt 126 19 19 100.00 15.08 26.21 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 60.08 45.73 51.93 Avg2. 23486 23575 20917 88.73 89.06 88.89 Current max chunk-based F1: 88.89 (iteration 8) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 9 Log-likelihood = -597187.903839 Norm (log-likelihood gradient vector) = 47564.519862 Norm (lambda vector) = 78.755947 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4855 4526 93.22 97.69 95.40 i-np 13660 13810 13143 95.17 96.22 95.69 e-np 12220 12267 11852 96.62 96.99 96.80 o 6349 6289 6036 95.98 95.07 95.52 e-vp 4768 4851 4559 93.98 95.62 94.79 i-vp 2602 2840 2497 87.92 95.96 91.77 e-adjp 384 342 270 78.95 70.31 74.38 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 767 581 75.75 70.68 73.13 i-advp 100 4 4 100.00 4.00 7.69 e-sbar 503 328 305 92.99 60.64 73.41 i-adjp 152 51 37 72.55 24.34 36.45 e-prt 126 47 46 97.87 36.51 53.18 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 54.05 42.20 47.40 Avg2. 46451 46451 43856 94.41 94.41 94.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12282 11193 91.13 91.60 91.36 pp 4633 4855 4487 92.42 96.85 94.58 vp 4768 4852 4338 89.41 90.98 90.19 sbar 503 328 300 91.46 59.64 72.20 adjp 384 342 218 63.74 56.77 60.06 advp 822 767 535 69.75 65.09 67.34 prt 126 47 46 97.87 36.51 53.18 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 59.58 49.74 54.22 Avg2. 23486 23473 21117 89.96 89.91 89.94 Current max chunk-based F1: 89.94 (iteration 9) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 10 Log-likelihood = -518688.050101 Norm (log-likelihood gradient vector) = 28783.031427 Norm (lambda vector) = 80.582401 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4802 4539 94.52 97.97 96.22 i-np 13660 13624 13111 96.23 95.98 96.11 e-np 12220 12285 11902 96.88 97.40 97.14 o 6349 6379 6105 95.70 96.16 95.93 e-vp 4768 4861 4600 94.63 96.48 95.54 i-vp 2602 2725 2483 91.12 95.43 93.22 e-adjp 384 395 298 75.44 77.60 76.51 i-pp 52 0 0 0.00 0.00 0.00 e-advp 822 817 628 76.87 76.40 76.63 i-advp 100 9 8 88.89 8.00 14.68 e-sbar 503 387 355 91.73 70.58 79.78 i-adjp 152 82 56 68.29 36.84 47.86 e-prt 126 85 83 97.65 65.87 78.67 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 53.40 45.74 49.27 Avg2. 46451 46451 44168 95.09 95.09 95.09 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12294 11295 91.87 92.43 92.15 pp 4633 4802 4500 93.71 97.13 95.39 vp 4768 4861 4411 90.74 92.51 91.62 sbar 503 387 349 90.18 69.38 78.43 adjp 384 395 248 62.78 64.58 63.67 advp 822 817 584 71.48 71.05 71.26 prt 126 85 83 97.65 65.87 78.67 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 59.84 55.30 57.48 Avg2. 23486 23641 21470 90.82 91.42 91.12 Current max chunk-based F1: 91.12 (iteration 10) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 11 Log-likelihood = -465962.273484 Norm (log-likelihood gradient vector) = 23450.585490 Norm (lambda vector) = 83.367155 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4821 4548 94.34 98.17 96.21 i-np 13660 14262 13335 93.50 97.62 95.52 e-np 12220 11912 11677 98.03 95.56 96.78 o 6349 6092 5916 97.11 93.18 95.10 e-vp 4768 4828 4617 95.63 96.83 96.23 i-vp 2602 2681 2478 92.43 95.23 93.81 e-adjp 384 410 308 75.12 80.21 77.58 i-pp 52 5 5 100.00 9.62 17.54 e-advp 822 800 638 79.75 77.62 78.67 i-advp 100 43 32 74.42 32.00 44.76 e-sbar 503 380 342 90.00 67.99 77.46 i-adjp 152 112 77 68.75 50.66 58.33 e-prt 126 105 102 97.14 80.95 88.31 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 57.81 48.78 52.91 Avg2. 46451 46451 44075 94.88 94.88 94.88 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11928 11032 92.49 90.28 91.37 pp 4633 4821 4514 93.63 97.43 95.49 vp 4768 4828 4438 91.92 93.08 92.50 sbar 503 380 336 88.42 66.80 76.10 adjp 384 410 268 65.37 69.79 67.51 advp 822 800 615 76.88 74.82 75.83 prt 126 105 102 97.14 80.95 88.31 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 60.58 57.31 58.90 Avg2. 23486 23272 21305 91.55 90.71 91.13 Current max chunk-based F1: 91.13 (iteration 11) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 12 Log-likelihood = -439009.882663 Norm (log-likelihood gradient vector) = 44916.031119 Norm (lambda vector) = 87.970197 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4821 4552 94.42 98.25 96.30 i-np 13660 13555 13130 96.86 96.12 96.49 e-np 12220 12275 11923 97.13 97.57 97.35 o 6349 6417 6140 95.68 96.71 96.19 e-vp 4768 4828 4623 95.75 96.96 96.35 i-vp 2602 2677 2479 92.60 95.27 93.92 e-adjp 384 405 307 75.80 79.95 77.82 i-pp 52 13 13 100.00 25.00 40.00 e-advp 822 791 638 80.66 77.62 79.11 i-advp 100 59 39 66.10 39.00 49.06 e-sbar 503 379 345 91.03 68.59 78.23 i-adjp 152 120 81 67.50 53.29 59.56 e-prt 126 111 108 97.30 85.71 91.14 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 57.54 50.50 53.79 Avg2. 46451 46451 44378 95.54 95.54 95.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12287 11407 92.84 93.35 93.09 pp 4633 4821 4526 93.88 97.69 95.75 vp 4768 4828 4448 92.13 93.29 92.71 sbar 503 379 338 89.18 67.20 76.64 adjp 384 405 269 66.42 70.05 68.19 advp 822 791 620 78.38 75.43 76.88 prt 126 111 108 97.30 85.71 91.14 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 61.01 58.27 59.61 Avg2. 23486 23622 21716 91.93 92.46 92.20 Current max chunk-based F1: 92.20 (iteration 12) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 13 Log-likelihood = -405736.578902 Norm (log-likelihood gradient vector) = 20322.804580 Norm (lambda vector) = 90.113170 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4808 4554 94.72 98.29 96.47 i-np 13660 13527 13127 97.04 96.10 96.57 e-np 12220 12313 11941 96.98 97.72 97.35 o 6349 6442 6154 95.53 96.93 96.22 e-vp 4768 4820 4625 95.95 97.00 96.47 i-vp 2602 2681 2485 92.69 95.50 94.08 e-adjp 384 381 302 79.27 78.65 78.95 i-pp 52 26 26 100.00 50.00 66.67 e-advp 822 781 636 81.43 77.37 79.35 i-advp 100 63 42 66.67 42.00 51.53 e-sbar 503 374 345 92.25 68.59 78.68 i-adjp 152 117 81 69.23 53.29 60.22 e-prt 126 118 112 94.92 88.89 91.80 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 57.83 52.02 54.77 Avg2. 46451 46451 44430 95.65 95.65 95.65 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12317 11442 92.90 93.63 93.26 pp 4633 4808 4541 94.45 98.01 96.20 vp 4768 4820 4449 92.30 93.31 92.80 sbar 503 374 338 90.37 67.20 77.08 adjp 384 381 266 69.82 69.27 69.54 advp 822 781 618 79.13 75.18 77.11 prt 126 118 112 94.92 88.89 91.80 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 61.39 58.55 59.94 Avg2. 23486 23599 21766 92.23 92.68 92.45 Current max chunk-based F1: 92.45 (iteration 13) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 14 Log-likelihood = -390600.126725 Norm (log-likelihood gradient vector) = 17975.581637 Norm (lambda vector) = 91.177120 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4773 4558 95.50 98.38 96.92 i-np 13660 13407 13066 97.46 95.65 96.55 e-np 12220 12401 11988 96.67 98.10 97.38 o 6349 6482 6171 95.20 97.20 96.19 e-vp 4768 4808 4631 96.32 97.13 96.72 i-vp 2602 2672 2493 93.30 95.81 94.54 e-adjp 384 359 295 82.17 76.82 79.41 i-pp 52 34 33 97.06 63.46 76.74 e-advp 822 787 653 82.97 79.44 81.17 i-advp 100 76 49 64.47 49.00 55.68 e-sbar 503 405 374 92.35 74.35 82.38 i-adjp 152 122 86 70.49 56.58 62.77 e-prt 126 125 118 94.40 93.65 94.02 i-sbar 12 0 0 0.00 0.00 0.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 57.92 53.78 55.77 Avg2. 46451 46451 44515 95.83 95.83 95.83 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12405 11503 92.73 94.13 93.43 pp 4633 4773 4550 95.33 98.21 96.75 vp 4768 4808 4476 93.09 93.88 93.48 sbar 503 405 364 89.88 72.37 80.18 adjp 384 359 264 73.54 68.75 71.06 advp 822 787 637 80.94 77.49 79.18 prt 126 125 118 94.40 93.65 94.02 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 61.99 59.85 60.90 Avg2. 23486 23662 21912 92.60 93.30 92.95 Current max chunk-based F1: 92.95 (iteration 14) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 15 Log-likelihood = -368284.501512 Norm (log-likelihood gradient vector) = 27511.569014 Norm (lambda vector) = 94.280868 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4699 4535 96.51 97.88 97.19 i-np 13660 13610 13182 96.86 96.50 96.68 e-np 12220 12298 11946 97.14 97.76 97.45 o 6349 6374 6120 96.02 96.39 96.20 e-vp 4768 4785 4628 96.72 97.06 96.89 i-vp 2602 2678 2495 93.17 95.89 94.51 e-adjp 384 348 291 83.62 75.78 79.51 i-pp 52 41 33 80.49 63.46 70.97 e-advp 822 811 664 81.87 80.78 81.32 i-advp 100 83 54 65.06 54.00 59.02 e-sbar 503 470 412 87.66 81.91 84.69 i-adjp 152 122 95 77.87 62.50 69.34 e-prt 126 127 120 94.49 95.24 94.86 i-sbar 12 2 1 50.00 8.33 14.29 i-conjp 24 2 2 100.00 8.33 15.38 e-conjp 16 1 1 100.00 6.25 11.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 69.87 55.90 62.11 Avg2. 46451 46451 44579 95.97 95.97 95.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12299 11477 93.32 93.92 93.62 pp 4633 4699 4519 96.17 97.54 96.85 vp 4768 4785 4473 93.48 93.81 93.65 sbar 503 470 400 85.11 79.52 82.22 adjp 384 348 267 76.72 69.53 72.95 advp 822 811 650 80.15 79.08 79.61 prt 126 127 120 94.49 95.24 94.86 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 1 1 100.00 6.25 11.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 71.94 61.49 66.31 Avg2. 23486 23540 21907 93.06 93.28 93.17 Current max chunk-based F1: 93.17 (iteration 15) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 16 Log-likelihood = -348418.606048 Norm (log-likelihood gradient vector) = 22012.682618 Norm (lambda vector) = 97.074394 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4709 4549 96.60 98.19 97.39 i-np 13660 13589 13195 97.10 96.60 96.85 e-np 12220 12289 11954 97.27 97.82 97.55 o 6349 6383 6136 96.13 96.65 96.39 e-vp 4768 4774 4630 96.98 97.11 97.04 i-vp 2602 2691 2502 92.98 96.16 94.54 e-adjp 384 344 290 84.30 75.52 79.67 i-pp 52 43 33 76.74 63.46 69.47 e-advp 822 827 673 81.38 81.87 81.63 i-advp 100 90 59 65.56 59.00 62.11 e-sbar 503 452 409 90.49 81.31 85.65 i-adjp 152 126 102 80.95 67.11 73.38 e-prt 126 128 120 93.75 95.24 94.49 i-sbar 12 3 1 33.33 8.33 13.33 i-conjp 24 2 2 100.00 8.33 15.38 e-conjp 16 1 1 100.00 6.25 11.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 69.18 56.45 62.17 Avg2. 46451 46451 44656 96.14 96.14 96.14 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12290 11513 93.68 94.21 93.95 pp 4633 4709 4532 96.24 97.82 97.02 vp 4768 4774 4477 93.78 93.90 93.84 sbar 503 452 396 87.61 78.73 82.93 adjp 384 344 272 79.07 70.83 74.73 advp 822 827 659 79.69 80.17 79.93 prt 126 128 120 93.75 95.24 94.49 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 1 1 100.00 6.25 11.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.38 61.72 66.62 Avg2. 23486 23525 21970 93.39 93.55 93.47 Current max chunk-based F1: 93.47 (iteration 16) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 17 Log-likelihood = -332091.636661 Norm (log-likelihood gradient vector) = 13291.063350 Norm (lambda vector) = 98.483884 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4715 4556 96.63 98.34 97.48 i-np 13660 13776 13317 96.67 97.49 97.08 e-np 12220 12173 11914 97.87 97.50 97.68 o 6349 6268 6104 97.38 96.14 96.76 e-vp 4768 4764 4633 97.25 97.17 97.21 i-vp 2602 2701 2515 93.11 96.66 94.85 e-adjp 384 351 296 84.33 77.08 80.54 i-pp 52 40 33 82.50 63.46 71.74 e-advp 822 855 688 80.47 83.70 82.05 i-advp 100 91 60 65.93 60.00 62.83 e-sbar 503 444 409 92.12 81.31 86.38 i-adjp 152 131 109 83.21 71.71 77.03 e-prt 126 129 120 93.02 95.24 94.12 i-sbar 12 4 2 50.00 16.67 25.00 i-conjp 24 6 6 100.00 25.00 40.00 e-conjp 16 3 3 100.00 18.75 31.58 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 70.52 58.81 64.14 Avg2. 46451 46451 44765 96.37 96.37 96.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12176 11490 94.37 94.03 94.20 pp 4633 4715 4542 96.33 98.04 97.18 vp 4768 4764 4486 94.16 94.09 94.13 sbar 503 444 397 89.41 78.93 83.84 adjp 384 351 281 80.06 73.18 76.46 advp 822 855 674 78.83 82.00 80.38 prt 126 129 120 93.02 95.24 94.12 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 3 3 100.00 18.75 31.58 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.62 63.42 67.71 Avg2. 23486 23437 21993 93.84 93.64 93.74 Current max chunk-based F1: 93.74 (iteration 17) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 18 Log-likelihood = -314020.121995 Norm (log-likelihood gradient vector) = 13389.021441 Norm (lambda vector) = 100.224946 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4726 4564 96.57 98.51 97.53 i-np 13660 13724 13300 96.91 97.36 97.14 e-np 12220 12196 11940 97.90 97.71 97.80 o 6349 6298 6124 97.24 96.46 96.85 e-vp 4768 4762 4629 97.21 97.08 97.15 i-vp 2602 2702 2515 93.08 96.66 94.83 e-adjp 384 356 303 85.11 78.91 81.89 i-pp 52 38 33 86.84 63.46 73.33 e-advp 822 854 695 81.38 84.55 82.94 i-advp 100 87 59 67.82 59.00 63.10 e-sbar 503 434 406 93.55 80.72 86.66 i-adjp 152 124 107 86.29 70.39 77.54 e-prt 126 129 120 93.02 95.24 94.12 i-sbar 12 4 2 50.00 16.67 25.00 i-conjp 24 12 12 100.00 50.00 66.67 e-conjp 16 5 5 100.00 31.25 47.62 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 71.15 60.70 65.51 Avg2. 46451 46451 44814 96.48 96.48 96.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12198 11534 94.56 94.39 94.47 pp 4633 4726 4554 96.36 98.29 97.32 vp 4768 4762 4480 94.08 93.96 94.02 sbar 503 434 394 90.78 78.33 84.10 adjp 384 356 289 81.18 75.26 78.11 advp 822 854 680 79.63 82.73 81.15 prt 126 129 120 93.02 95.24 94.12 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 6 5 83.33 31.25 45.45 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 71.29 64.94 67.97 Avg2. 23486 23465 22056 94.00 93.91 93.95 Current max chunk-based F1: 93.95 (iteration 18) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 19 Log-likelihood = -300226.780902 Norm (log-likelihood gradient vector) = 14289.099488 Norm (lambda vector) = 102.490004 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4730 4569 96.60 98.62 97.60 i-np 13660 13825 13361 96.64 97.81 97.22 e-np 12220 12131 11921 98.27 97.55 97.91 o 6349 6243 6099 97.69 96.06 96.87 e-vp 4768 4770 4641 97.30 97.34 97.32 i-vp 2602 2704 2524 93.34 97.00 95.14 e-adjp 384 374 311 83.16 80.99 82.06 i-pp 52 36 33 91.67 63.46 75.00 e-advp 822 827 690 83.43 83.94 83.69 i-advp 100 80 59 73.75 59.00 65.56 e-sbar 503 439 416 94.76 82.70 88.32 i-adjp 152 131 111 84.73 73.03 78.45 e-prt 126 129 120 93.02 95.24 94.12 i-sbar 12 6 4 66.67 33.33 44.44 i-conjp 24 18 16 88.89 66.67 76.19 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.00 63.64 67.56 Avg2. 46451 46451 44883 96.62 96.62 96.62 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12132 11527 95.01 94.33 94.67 pp 4633 4730 4561 96.43 98.45 97.43 vp 4768 4770 4498 94.30 94.34 94.32 sbar 503 439 406 92.48 80.72 86.20 adjp 384 374 297 79.41 77.34 78.36 advp 822 827 676 81.74 82.24 81.99 prt 126 129 120 93.02 95.24 94.12 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.13 67.26 69.61 Avg2. 23486 23410 22093 94.37 94.07 94.22 Current max chunk-based F1: 94.22 (iteration 19) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 20 Log-likelihood = -280731.329684 Norm (log-likelihood gradient vector) = 15805.949659 Norm (lambda vector) = 106.869016 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4689 4553 97.10 98.27 97.68 i-np 13660 13287 13034 98.10 95.42 96.74 e-np 12220 12409 12026 96.91 98.41 97.66 o 6349 6511 6201 95.24 97.67 96.44 e-vp 4768 4773 4643 97.28 97.38 97.33 i-vp 2602 2718 2533 93.19 97.35 95.23 e-adjp 384 378 314 83.07 81.77 82.41 i-pp 52 32 31 96.88 59.62 73.81 e-advp 822 804 676 84.08 82.24 83.15 i-advp 100 72 57 79.17 57.00 66.28 e-sbar 503 475 442 93.05 87.87 90.39 i-adjp 152 137 114 83.21 75.00 78.89 e-prt 126 130 121 93.08 96.03 94.53 i-sbar 12 7 5 71.43 41.67 52.63 i-conjp 24 20 18 90.00 75.00 81.82 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.59 64.85 68.50 Avg2. 46451 46451 44777 96.40 96.40 96.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12409 11595 93.44 94.89 94.16 pp 4633 4689 4544 96.91 98.08 97.49 vp 4768 4773 4498 94.24 94.34 94.29 sbar 503 475 433 91.16 86.08 88.55 adjp 384 378 302 79.89 78.65 79.27 advp 822 804 664 82.59 80.78 81.67 prt 126 130 121 93.08 96.03 94.53 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.13 68.51 70.27 Avg2. 23486 23668 22166 93.65 94.38 94.02 Current max chunk-based F1: 94.22 (iteration 19) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 21 Log-likelihood = -271794.139994 Norm (log-likelihood gradient vector) = 26429.650368 Norm (lambda vector) = 112.430036 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4697 4561 97.10 98.45 97.77 i-np 13660 13576 13244 97.55 96.95 97.25 e-np 12220 12273 11996 97.74 98.17 97.95 o 6349 6388 6175 96.67 97.26 96.96 e-vp 4768 4771 4642 97.30 97.36 97.33 i-vp 2602 2724 2536 93.10 97.46 95.23 e-adjp 384 376 313 83.24 81.51 82.37 i-pp 52 33 32 96.97 61.54 75.29 e-advp 822 776 665 85.70 80.90 83.23 i-advp 100 72 57 79.17 57.00 66.28 e-sbar 503 466 440 94.42 87.48 90.82 i-adjp 152 134 112 83.58 73.68 78.32 e-prt 126 129 121 93.80 96.03 94.90 i-sbar 12 7 5 71.43 41.67 52.63 i-conjp 24 20 18 90.00 75.00 81.82 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.89 64.84 68.63 Avg2. 46451 46451 44926 96.72 96.72 96.72 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12273 11614 94.63 95.04 94.84 pp 4633 4697 4553 96.93 98.27 97.60 vp 4768 4771 4498 94.28 94.34 94.31 sbar 503 466 431 92.49 85.69 88.96 adjp 384 376 301 80.05 78.39 79.21 advp 822 776 653 84.15 79.44 81.73 prt 126 129 121 93.80 96.03 94.90 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.63 68.34 70.42 Avg2. 23486 23498 22180 94.39 94.44 94.42 Current max chunk-based F1: 94.42 (iteration 21) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 22 Log-likelihood = -257655.090407 Norm (log-likelihood gradient vector) = 10257.075649 Norm (lambda vector) = 111.569412 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4692 4560 97.19 98.42 97.80 i-np 13660 13712 13308 97.05 97.42 97.24 e-np 12220 12208 11964 98.00 97.91 97.95 o 6349 6327 6146 97.14 96.80 96.97 e-vp 4768 4773 4645 97.32 97.42 97.37 i-vp 2602 2721 2535 93.16 97.43 95.25 e-adjp 384 374 314 83.96 81.77 82.85 i-pp 52 34 33 97.06 63.46 76.74 e-advp 822 764 662 86.65 80.54 83.48 i-advp 100 73 58 79.45 58.00 67.05 e-sbar 503 472 443 93.86 88.07 90.87 i-adjp 152 136 112 82.35 73.68 77.78 e-prt 126 128 120 93.75 95.24 94.49 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 20 18 90.00 75.00 81.82 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.47 64.95 68.51 Avg2. 46451 46451 44932 96.73 96.73 96.73 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12208 11581 94.86 94.77 94.82 pp 4633 4692 4553 97.04 98.27 97.65 vp 4768 4773 4500 94.28 94.38 94.33 sbar 503 472 433 91.74 86.08 88.82 adjp 384 374 301 80.48 78.39 79.42 advp 822 764 649 84.95 78.95 81.84 prt 126 128 120 93.75 95.24 94.49 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.71 68.23 70.40 Avg2. 23486 23421 22146 94.56 94.29 94.43 Current max chunk-based F1: 94.43 (iteration 22) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 23 Log-likelihood = -252667.528537 Norm (log-likelihood gradient vector) = 8515.370072 Norm (lambda vector) = 111.833638 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4682 4555 97.29 98.32 97.80 i-np 13660 13771 13325 96.76 97.55 97.15 e-np 12220 12183 11945 98.05 97.75 97.90 o 6349 6302 6126 97.21 96.49 96.85 e-vp 4768 4775 4648 97.34 97.48 97.41 i-vp 2602 2710 2531 93.39 97.27 95.29 e-adjp 384 370 315 85.14 82.03 83.55 i-pp 52 34 33 97.06 63.46 76.74 e-advp 822 767 662 86.31 80.54 83.32 i-advp 100 73 58 79.45 58.00 67.05 e-sbar 503 482 447 92.74 88.87 90.76 i-adjp 152 136 113 83.09 74.34 78.47 e-prt 126 129 121 93.80 96.03 94.90 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 20 18 90.00 75.00 81.82 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.51 65.05 68.58 Avg2. 46451 46451 44911 96.68 96.68 96.68 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12183 11549 94.80 94.51 94.65 pp 4633 4682 4548 97.14 98.17 97.65 vp 4768 4775 4506 94.37 94.51 94.44 sbar 503 482 437 90.66 86.88 88.73 adjp 384 370 302 81.62 78.65 80.11 advp 822 767 649 84.62 78.95 81.69 prt 126 129 121 93.80 96.03 94.90 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.70 68.39 70.48 Avg2. 23486 23398 22121 94.54 94.19 94.36 Current max chunk-based F1: 94.43 (iteration 22) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 24 Log-likelihood = -247838.125072 Norm (log-likelihood gradient vector) = 10430.810744 Norm (lambda vector) = 112.877834 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4683 4559 97.35 98.40 97.87 i-np 13660 13744 13329 96.98 97.58 97.28 e-np 12220 12202 11965 98.06 97.91 97.99 o 6349 6320 6138 97.12 96.68 96.90 e-vp 4768 4770 4650 97.48 97.53 97.50 i-vp 2602 2678 2517 93.99 96.73 95.34 e-adjp 384 371 319 85.98 83.07 84.50 i-pp 52 36 34 94.44 65.38 77.27 e-advp 822 778 676 86.89 82.24 84.50 i-advp 100 77 60 77.92 60.00 67.80 e-sbar 503 484 452 93.39 89.86 91.59 i-adjp 152 142 115 80.99 75.66 78.23 e-prt 126 129 121 93.80 96.03 94.90 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 20 18 90.00 75.00 81.82 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.34 65.50 68.75 Avg2. 46451 46451 44967 96.81 96.81 96.81 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12202 11585 94.94 94.80 94.87 pp 4633 4683 4553 97.22 98.27 97.75 vp 4768 4770 4516 94.68 94.71 94.69 sbar 503 484 442 91.32 87.87 89.56 adjp 384 371 305 82.21 79.43 80.79 advp 822 778 664 85.35 80.78 83.00 prt 126 129 121 93.80 96.03 94.90 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.95 68.82 70.82 Avg2. 23486 23427 22195 94.74 94.50 94.62 Current max chunk-based F1: 94.62 (iteration 24) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 25 Log-likelihood = -237412.454341 Norm (log-likelihood gradient vector) = 8843.827694 Norm (lambda vector) = 116.470295 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4632 4535 97.91 97.88 97.90 i-np 13660 14196 13432 94.62 98.33 96.44 e-np 12220 11984 11796 98.43 96.53 97.47 o 6349 6089 5954 97.78 93.78 95.74 e-vp 4768 4740 4638 97.85 97.27 97.56 i-vp 2602 2668 2518 94.38 96.77 95.56 e-adjp 384 361 321 88.92 83.59 86.17 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 833 715 85.83 86.98 86.40 i-advp 100 74 65 87.84 65.00 74.71 e-sbar 503 530 468 88.30 93.04 90.61 i-adjp 152 142 116 81.69 76.32 78.91 e-prt 126 129 122 94.57 96.83 95.69 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 18 18 100.00 75.00 85.71 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.00 66.03 69.34 Avg2. 46451 46451 44746 96.33 96.33 96.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 11986 11375 94.90 93.09 93.98 pp 4633 4632 4529 97.78 97.76 97.77 vp 4768 4740 4508 95.11 94.55 94.83 sbar 503 530 458 86.42 91.05 88.67 adjp 384 361 308 85.32 80.21 82.68 advp 822 833 706 84.75 85.89 85.32 prt 126 129 122 94.57 96.83 95.69 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.88 69.56 71.66 Avg2. 23486 23220 22015 94.81 93.74 94.27 Current max chunk-based F1: 94.62 (iteration 24) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 26 Log-likelihood = -232366.971386 Norm (log-likelihood gradient vector) = 27654.482229 Norm (lambda vector) = 123.646243 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4681 4563 97.48 98.49 97.98 i-np 13660 13698 13334 97.34 97.61 97.48 e-np 12220 12222 11985 98.06 98.08 98.07 o 6349 6327 6143 97.09 96.76 96.92 e-vp 4768 4751 4649 97.85 97.50 97.68 i-vp 2602 2661 2518 94.63 96.77 95.69 e-adjp 384 360 319 88.61 83.07 85.75 i-pp 52 37 34 91.89 65.38 76.40 e-advp 822 845 720 85.21 87.59 86.38 i-advp 100 79 67 84.81 67.00 74.86 e-sbar 503 492 458 93.09 91.05 92.06 i-adjp 152 141 113 80.14 74.34 77.13 e-prt 126 128 121 94.53 96.03 95.28 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 14 14 100.00 58.33 73.68 e-conjp 16 7 7 100.00 43.75 60.87 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.16 64.67 68.66 Avg2. 46451 46451 45050 96.98 96.98 96.98 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12222 11633 95.18 95.20 95.19 pp 4633 4681 4557 97.35 98.36 97.85 vp 4768 4751 4524 95.22 94.88 95.05 sbar 503 492 448 91.06 89.07 90.05 adjp 384 360 304 84.44 79.17 81.72 advp 822 845 711 84.14 86.50 85.30 prt 126 128 121 94.53 96.03 95.28 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 7 7 100.00 43.75 60.87 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.19 68.29 71.12 Avg2. 23486 23486 22305 94.97 94.97 94.97 Current max chunk-based F1: 94.97 (iteration 26) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 27 Log-likelihood = -220998.534547 Norm (log-likelihood gradient vector) = 8178.700224 Norm (lambda vector) = 124.127279 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4689 4569 97.44 98.62 98.03 i-np 13660 13665 13318 97.46 97.50 97.48 e-np 12220 12236 11994 98.02 98.15 98.09 o 6349 6343 6153 97.00 96.91 96.96 e-vp 4768 4750 4647 97.83 97.46 97.65 i-vp 2602 2670 2520 94.38 96.85 95.60 e-adjp 384 362 320 88.40 83.33 85.79 i-pp 52 37 34 91.89 65.38 76.40 e-advp 822 837 715 85.42 86.98 86.20 i-advp 100 80 68 85.00 68.00 75.56 e-sbar 503 484 454 93.80 90.26 92.00 i-adjp 152 141 113 80.14 74.34 77.13 e-prt 126 128 121 94.53 96.03 95.28 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 14 14 100.00 58.33 73.68 e-conjp 16 7 7 100.00 43.75 60.87 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.19 64.68 68.67 Avg2. 46451 46451 45052 96.99 96.99 96.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12236 11643 95.15 95.28 95.22 pp 4633 4689 4563 97.31 98.49 97.90 vp 4768 4750 4520 95.16 94.80 94.98 sbar 503 484 444 91.74 88.27 89.97 adjp 384 362 305 84.25 79.43 81.77 advp 822 837 707 84.47 86.01 85.23 prt 126 128 121 94.53 96.03 95.28 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 7 7 100.00 43.75 60.87 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.26 68.21 71.10 Avg2. 23486 23493 22310 94.96 94.99 94.98 Current max chunk-based F1: 94.98 (iteration 27) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 28 Log-likelihood = -217309.296151 Norm (log-likelihood gradient vector) = 6309.401098 Norm (lambda vector) = 123.895974 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4699 4575 97.36 98.75 98.05 i-np 13660 13558 13265 97.84 97.11 97.47 e-np 12220 12279 12013 97.83 98.31 98.07 o 6349 6395 6180 96.64 97.34 96.99 e-vp 4768 4750 4647 97.83 97.46 97.65 i-vp 2602 2685 2525 94.04 97.04 95.52 e-adjp 384 365 325 89.04 84.64 86.78 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 825 711 86.18 86.50 86.34 i-advp 100 85 68 80.00 68.00 73.51 e-sbar 503 481 453 94.18 90.06 92.07 i-adjp 152 140 114 81.43 75.00 78.08 e-prt 126 128 121 94.53 96.03 95.28 i-sbar 12 8 5 62.50 41.67 50.00 i-conjp 24 10 10 100.00 41.67 58.82 e-conjp 16 5 5 100.00 31.25 47.62 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 72.94 63.31 67.79 Avg2. 46451 46451 45051 96.99 96.99 96.99 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12279 11668 95.02 95.48 95.25 pp 4633 4699 4568 97.21 98.60 97.90 vp 4768 4750 4519 95.14 94.78 94.96 sbar 503 481 443 92.10 88.07 90.04 adjp 384 365 310 84.93 80.73 82.78 advp 822 825 703 85.21 85.52 85.37 prt 126 128 121 94.53 96.03 95.28 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 5 5 100.00 31.25 47.62 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.41 67.05 70.54 Avg2. 23486 23532 22337 94.92 95.11 95.01 Current max chunk-based F1: 95.01 (iteration 28) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 29 Log-likelihood = -209451.364298 Norm (log-likelihood gradient vector) = 7741.504481 Norm (lambda vector) = 125.266754 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4698 4573 97.34 98.70 98.02 i-np 13660 13746 13359 97.18 97.80 97.49 e-np 12220 12180 11969 98.27 97.95 98.11 o 6349 6314 6147 97.36 96.82 97.09 e-vp 4768 4750 4643 97.75 97.38 97.56 i-vp 2602 2676 2522 94.25 96.93 95.57 e-adjp 384 376 324 86.17 84.38 85.26 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 821 706 85.99 85.89 85.94 i-advp 100 87 67 77.01 67.00 71.66 e-sbar 503 477 449 94.13 89.26 91.63 i-adjp 152 142 114 80.28 75.00 77.55 e-prt 126 129 122 94.57 96.83 95.69 i-sbar 12 8 6 75.00 50.00 60.00 i-conjp 24 6 6 100.00 25.00 40.00 e-conjp 16 3 3 100.00 18.75 31.58 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.24 62.15 67.24 Avg2. 46451 46451 45044 96.97 96.97 96.97 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12181 11619 95.39 95.08 95.23 pp 4633 4698 4566 97.19 98.55 97.87 vp 4768 4750 4520 95.16 94.80 94.98 sbar 503 477 441 92.45 87.67 90.00 adjp 384 376 309 82.18 80.47 81.32 advp 822 821 697 84.90 84.79 84.84 prt 126 129 122 94.57 96.83 95.69 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 3 3 100.00 18.75 31.58 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.18 65.69 69.68 Avg2. 23486 23435 22277 95.06 94.85 94.96 Current max chunk-based F1: 95.01 (iteration 28) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 30 Log-likelihood = -201863.297358 Norm (log-likelihood gradient vector) = 9600.331773 Norm (lambda vector) = 128.273400 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4697 4573 97.36 98.70 98.03 i-np 13660 13588 13274 97.69 97.17 97.43 e-np 12220 12258 12008 97.96 98.27 98.11 o 6349 6384 6177 96.76 97.29 97.02 e-vp 4768 4754 4644 97.69 97.40 97.54 i-vp 2602 2695 2532 93.95 97.31 95.60 e-adjp 384 385 328 85.19 85.42 85.31 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 804 694 86.32 84.43 85.36 i-advp 100 89 67 75.28 67.00 70.90 e-sbar 503 480 450 93.75 89.46 91.56 i-adjp 152 142 114 80.28 75.00 77.55 e-prt 126 129 122 94.57 96.83 95.69 i-sbar 12 8 6 75.00 50.00 60.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 63.06 59.98 61.48 Avg2. 46451 46451 45023 96.93 96.93 96.93 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12259 11661 95.12 95.43 95.27 pp 4633 4697 4565 97.19 98.53 97.86 vp 4768 4754 4524 95.16 94.88 95.02 sbar 503 480 442 92.08 87.87 89.93 adjp 384 385 312 81.04 81.25 81.14 advp 822 804 685 85.20 83.33 84.26 prt 126 129 122 94.57 96.83 95.69 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 64.04 63.81 63.92 Avg2. 23486 23508 22311 94.91 95.00 94.95 Current max chunk-based F1: 95.01 (iteration 28) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 31 Log-likelihood = -198031.527587 Norm (log-likelihood gradient vector) = 8620.257888 Norm (lambda vector) = 131.019907 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4696 4574 97.40 98.73 98.06 i-np 13660 13694 13346 97.46 97.70 97.58 e-np 12220 12213 11997 98.23 98.18 98.20 o 6349 6334 6162 97.28 97.05 97.17 e-vp 4768 4750 4645 97.79 97.42 97.60 i-vp 2602 2696 2534 93.99 97.39 95.66 e-adjp 384 384 329 85.68 85.68 85.68 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 798 691 86.59 84.06 85.31 i-advp 100 88 67 76.14 67.00 71.28 e-sbar 503 481 452 93.97 89.86 91.87 i-adjp 152 141 114 80.85 75.00 77.82 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 8 6 75.00 50.00 60.00 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 63.22 60.05 61.60 Avg2. 46451 46451 45074 97.04 97.04 97.04 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12214 11664 95.50 95.45 95.47 pp 4633 4696 4566 97.23 98.55 97.89 vp 4768 4750 4527 95.31 94.95 95.13 sbar 503 481 444 92.31 88.27 90.24 adjp 384 384 313 81.51 81.51 81.51 advp 822 798 682 85.46 82.97 84.20 prt 126 130 123 94.62 97.62 96.09 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 64.19 63.93 64.06 Avg2. 23486 23453 22319 95.16 95.03 95.10 Current max chunk-based F1: 95.10 (iteration 31) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 32 Log-likelihood = -193771.086151 Norm (log-likelihood gradient vector) = 6028.222187 Norm (lambda vector) = 131.844488 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4673 4565 97.69 98.53 98.11 i-np 13660 13710 13366 97.49 97.85 97.67 e-np 12220 12207 11998 98.29 98.18 98.24 o 6349 6335 6167 97.35 97.13 97.24 e-vp 4768 4755 4649 97.77 97.50 97.64 i-vp 2602 2671 2526 94.57 97.08 95.81 e-adjp 384 377 327 86.74 85.16 85.94 i-pp 52 36 34 94.44 65.38 77.27 e-advp 822 820 706 86.10 85.89 85.99 i-advp 100 88 68 77.27 68.00 72.34 e-sbar 503 500 459 91.80 91.25 91.53 i-adjp 152 139 115 82.73 75.66 79.04 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 10 7 70.00 58.33 63.64 i-conjp 24 0 0 0.00 0.00 0.00 e-conjp 16 0 0 0.00 0.00 0.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 63.34 60.68 61.98 Avg2. 46451 46451 45110 97.11 97.11 97.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12208 11675 95.63 95.54 95.59 pp 4633 4673 4557 97.52 98.36 97.94 vp 4768 4755 4534 95.35 95.09 95.22 sbar 503 500 451 90.20 89.66 89.93 adjp 384 377 314 83.29 81.77 82.52 advp 822 820 698 85.12 84.91 85.02 prt 126 130 123 94.62 97.62 96.09 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 0 0 0.00 0.00 0.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 64.17 64.30 64.23 Avg2. 23486 23463 22352 95.26 95.17 95.22 Current max chunk-based F1: 95.22 (iteration 32) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 33 Log-likelihood = -187081.802582 Norm (log-likelihood gradient vector) = 5923.422297 Norm (lambda vector) = 134.794291 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4703 4581 97.41 98.88 98.14 i-np 13660 13754 13379 97.27 97.94 97.61 e-np 12220 12194 11989 98.32 98.11 98.21 o 6349 6314 6152 97.43 96.90 97.16 e-vp 4768 4751 4648 97.83 97.48 97.66 i-vp 2602 2660 2522 94.81 96.93 95.86 e-adjp 384 366 323 88.25 84.11 86.13 i-pp 52 36 34 94.44 65.38 77.27 e-advp 822 835 715 85.63 86.98 86.30 i-advp 100 88 68 77.27 68.00 72.34 e-sbar 503 473 448 94.71 89.07 91.80 i-adjp 152 138 113 81.88 74.34 77.93 e-prt 126 129 122 94.57 96.83 95.69 i-sbar 12 7 5 71.43 41.67 52.63 i-conjp 24 2 2 100.00 8.33 15.38 e-conjp 16 1 1 100.00 6.25 11.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.56 60.36 66.31 Avg2. 46451 46451 45102 97.10 97.10 97.10 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12195 11661 95.62 95.43 95.52 pp 4633 4703 4573 97.24 98.70 97.96 vp 4768 4751 4536 95.47 95.13 95.30 sbar 503 473 439 92.81 87.28 89.96 adjp 384 366 309 84.43 80.47 82.40 advp 822 835 708 84.79 86.13 85.46 prt 126 129 122 94.57 96.83 95.69 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 1 1 100.00 6.25 11.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.49 64.62 69.21 Avg2. 23486 23453 22349 95.29 95.16 95.23 Current max chunk-based F1: 95.23 (iteration 33) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 34 Log-likelihood = -180937.070918 Norm (log-likelihood gradient vector) = 7943.423637 Norm (lambda vector) = 137.652623 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4683 4573 97.65 98.70 98.18 i-np 13660 13681 13345 97.54 97.69 97.62 e-np 12220 12233 12007 98.15 98.26 98.20 o 6349 6357 6168 97.03 97.15 97.09 e-vp 4768 4750 4649 97.87 97.50 97.69 i-vp 2602 2652 2517 94.91 96.73 95.81 e-adjp 384 362 321 88.67 83.59 86.06 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 840 719 85.60 87.47 86.52 i-advp 100 85 68 80.00 68.00 73.51 e-sbar 503 485 453 93.40 90.06 91.70 i-adjp 152 138 113 81.88 74.34 77.93 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 7 5 71.43 41.67 52.63 i-conjp 24 8 8 100.00 33.33 50.00 e-conjp 16 4 4 100.00 25.00 40.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.80 62.67 67.78 Avg2. 46451 46451 45108 97.11 97.11 97.11 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11686 95.52 95.63 95.58 pp 4633 4683 4566 97.50 98.55 98.02 vp 4768 4750 4535 95.47 95.11 95.29 sbar 503 485 444 91.55 88.27 89.88 adjp 384 362 307 84.81 79.95 82.31 advp 822 840 713 84.88 86.74 85.80 prt 126 131 124 94.66 98.41 96.50 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 4 4 100.00 25.00 40.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.44 66.77 70.39 Avg2. 23486 23489 22379 95.27 95.29 95.28 Current max chunk-based F1: 95.28 (iteration 34) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 35 Log-likelihood = -173780.234869 Norm (log-likelihood gradient vector) = 5676.789101 Norm (lambda vector) = 140.451696 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4571 97.98 98.66 98.32 i-np 13660 13787 13404 97.22 98.13 97.67 e-np 12220 12192 11993 98.37 98.14 98.25 o 6349 6320 6156 97.41 96.96 97.18 e-vp 4768 4748 4647 97.87 97.46 97.67 i-vp 2602 2665 2521 94.60 96.89 95.73 e-adjp 384 349 317 90.83 82.55 86.49 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 821 713 86.85 86.74 86.79 i-advp 100 79 66 83.54 66.00 73.74 e-sbar 503 497 463 93.16 92.05 92.60 i-adjp 152 127 111 87.40 73.03 79.57 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 9 6 66.67 50.00 57.14 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.34 65.85 69.39 Avg2. 46451 46451 45150 97.20 97.20 97.20 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11681 95.80 95.59 95.69 pp 4633 4665 4564 97.83 98.51 98.17 vp 4768 4748 4531 95.43 95.03 95.23 sbar 503 497 454 91.35 90.26 90.80 adjp 384 349 305 87.39 79.43 83.22 advp 822 821 706 85.99 85.89 85.94 prt 126 131 124 94.66 98.41 96.50 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.73 69.31 71.45 Avg2. 23486 23413 22373 95.56 95.26 95.41 Current max chunk-based F1: 95.41 (iteration 35) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 36 Log-likelihood = -166544.924761 Norm (log-likelihood gradient vector) = 8248.371559 Norm (lambda vector) = 142.822510 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4683 4577 97.74 98.79 98.26 i-np 13660 13612 13316 97.83 97.48 97.65 e-np 12220 12266 12026 98.04 98.41 98.23 o 6349 6399 6194 96.80 97.56 97.18 e-vp 4768 4747 4647 97.89 97.46 97.68 i-vp 2602 2673 2531 94.69 97.27 95.96 e-adjp 384 365 322 88.22 83.85 85.98 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 805 707 87.83 86.01 86.91 i-advp 100 77 67 87.01 67.00 75.71 e-sbar 503 486 457 94.03 90.85 92.42 i-adjp 152 136 114 83.82 75.00 79.17 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 10 7 70.00 58.33 63.64 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.44 66.42 69.75 Avg2. 46451 46451 45147 97.19 97.19 97.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12267 11717 95.52 95.88 95.70 pp 4633 4683 4570 97.59 98.64 98.11 vp 4768 4747 4536 95.56 95.13 95.34 sbar 503 486 449 92.39 89.26 90.80 adjp 384 365 309 84.66 80.47 82.51 advp 822 805 701 87.08 85.28 86.17 prt 126 131 124 94.66 98.41 96.50 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.63 69.31 71.41 Avg2. 23486 23493 22414 95.41 95.44 95.42 Current max chunk-based F1: 95.42 (iteration 36) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 37 Log-likelihood = -161085.342838 Norm (log-likelihood gradient vector) = 6969.902711 Norm (lambda vector) = 144.237172 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4676 4575 97.84 98.75 98.29 i-np 13660 13690 13356 97.56 97.77 97.67 e-np 12220 12226 12007 98.21 98.26 98.23 o 6349 6363 6176 97.06 97.28 97.17 e-vp 4768 4748 4646 97.85 97.44 97.65 i-vp 2602 2679 2532 94.51 97.31 95.89 e-adjp 384 371 325 87.60 84.64 86.09 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 793 701 88.40 85.28 86.81 i-advp 100 78 68 87.18 68.00 76.40 e-sbar 503 489 459 93.87 91.25 92.54 i-adjp 152 137 114 83.21 75.00 78.89 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 10 7 70.00 58.33 63.64 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.22 66.49 70.14 Avg2. 46451 46451 45148 97.19 97.19 97.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11693 95.63 95.69 95.66 pp 4633 4676 4568 97.69 98.60 98.14 vp 4768 4748 4532 95.45 95.05 95.25 sbar 503 489 451 92.23 89.66 90.93 adjp 384 371 312 84.10 81.25 82.65 advp 822 793 696 87.77 84.67 86.19 prt 126 132 124 93.94 98.41 96.12 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.68 69.33 71.91 Avg2. 23486 23444 22384 95.48 95.31 95.39 Current max chunk-based F1: 95.42 (iteration 36) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 38 Log-likelihood = -158675.782687 Norm (log-likelihood gradient vector) = 4880.666460 Norm (lambda vector) = 144.181067 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4674 4575 97.88 98.75 98.31 i-np 13660 13691 13360 97.58 97.80 97.69 e-np 12220 12222 12008 98.25 98.27 98.26 o 6349 6354 6177 97.21 97.29 97.25 e-vp 4768 4751 4644 97.75 97.40 97.57 i-vp 2602 2675 2526 94.43 97.08 95.74 e-adjp 384 379 326 86.02 84.90 85.45 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 793 698 88.02 84.91 86.44 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 492 459 93.29 91.25 92.26 i-adjp 152 136 114 83.82 75.00 79.17 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 11 8 72.73 66.67 69.57 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.17 66.99 70.40 Avg2. 46451 46451 45147 97.19 97.19 97.19 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11694 95.67 95.70 95.68 pp 4633 4674 4568 97.73 98.60 98.16 vp 4768 4751 4529 95.33 94.99 95.16 sbar 503 492 453 92.07 90.06 91.06 adjp 384 379 313 82.59 81.51 82.04 advp 822 793 694 87.52 84.43 85.94 prt 126 132 124 93.94 98.41 96.12 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.48 69.37 71.84 Avg2. 23486 23452 22383 95.44 95.30 95.37 Current max chunk-based F1: 95.42 (iteration 36) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 39 Log-likelihood = -156156.625677 Norm (log-likelihood gradient vector) = 4210.079610 Norm (lambda vector) = 144.651469 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4677 4576 97.84 98.77 98.30 i-np 13660 13834 13421 97.01 98.25 97.63 e-np 12220 12155 11971 98.49 97.96 98.22 o 6349 6273 6135 97.80 96.63 97.21 e-vp 4768 4750 4646 97.81 97.44 97.63 i-vp 2602 2674 2529 94.58 97.19 95.87 e-adjp 384 382 327 85.60 85.16 85.38 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 794 700 88.16 85.16 86.63 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 485 455 93.81 90.46 92.11 i-adjp 152 136 114 83.82 75.00 79.17 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.06 67.57 70.67 Avg2. 46451 46451 45137 97.17 97.17 97.17 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12155 11647 95.82 95.31 95.57 pp 4633 4677 4569 97.69 98.62 98.15 vp 4768 4750 4535 95.47 95.11 95.29 sbar 503 485 450 92.78 89.46 91.09 adjp 384 382 315 82.46 82.03 82.25 advp 822 794 695 87.53 84.55 86.01 prt 126 133 124 93.23 98.41 95.75 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.50 69.35 71.83 Avg2. 23486 23384 22343 95.55 95.13 95.34 Current max chunk-based F1: 95.42 (iteration 36) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 40 Log-likelihood = -152188.601739 Norm (log-likelihood gradient vector) = 8661.925735 Norm (lambda vector) = 147.006996 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4679 4575 97.78 98.75 98.26 i-np 13660 13767 13411 97.41 98.18 97.79 e-np 12220 12188 11998 98.44 98.18 98.31 o 6349 6303 6160 97.73 97.02 97.38 e-vp 4768 4747 4648 97.91 97.48 97.70 i-vp 2602 2667 2531 94.90 97.27 96.07 e-adjp 384 384 328 85.42 85.42 85.42 i-pp 52 39 36 92.31 69.23 79.12 e-advp 822 804 704 87.56 85.64 86.59 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 485 454 93.61 90.26 91.90 i-adjp 152 134 112 83.58 73.68 78.32 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 73.86 67.56 70.57 Avg2. 46451 46451 45184 97.27 97.27 97.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12188 11700 96.00 95.74 95.87 pp 4633 4679 4568 97.63 98.60 98.11 vp 4768 4747 4544 95.72 95.30 95.51 sbar 503 485 449 92.58 89.26 90.89 adjp 384 384 313 81.51 81.51 81.51 advp 822 804 698 86.82 84.91 85.85 prt 126 133 124 93.23 98.41 95.75 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.35 69.37 71.78 Avg2. 23486 23428 22404 95.63 95.39 95.51 Current max chunk-based F1: 95.51 (iteration 40) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 41 Log-likelihood = -149010.283377 Norm (log-likelihood gradient vector) = 5182.440522 Norm (lambda vector) = 148.789066 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4679 4574 97.76 98.73 98.24 i-np 13660 13680 13365 97.70 97.84 97.77 e-np 12220 12248 12027 98.20 98.42 98.31 o 6349 6356 6182 97.26 97.37 97.32 e-vp 4768 4744 4649 98.00 97.50 97.75 i-vp 2602 2663 2530 95.01 97.23 96.11 e-adjp 384 372 323 86.83 84.11 85.45 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 811 708 87.30 86.13 86.71 i-advp 100 85 69 81.18 69.00 74.59 e-sbar 503 478 452 94.56 89.86 92.15 i-adjp 152 128 112 87.50 73.68 80.00 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.21 67.46 70.68 Avg2. 46451 46451 45184 97.27 97.27 97.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12248 11727 95.75 95.97 95.86 pp 4633 4679 4567 97.61 98.58 98.09 vp 4768 4744 4548 95.87 95.39 95.63 sbar 503 478 446 93.31 88.67 90.93 adjp 384 372 312 83.87 81.25 82.54 advp 822 811 701 86.44 85.28 85.85 prt 126 133 124 93.23 98.41 95.75 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.61 69.35 71.88 Avg2. 23486 23473 22433 95.57 95.52 95.54 Current max chunk-based F1: 95.54 (iteration 41) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 42 Log-likelihood = -145559.086137 Norm (log-likelihood gradient vector) = 3951.219568 Norm (lambda vector) = 151.084333 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4674 4574 97.86 98.73 98.29 i-np 13660 13655 13351 97.77 97.74 97.76 e-np 12220 12265 12037 98.14 98.50 98.32 o 6349 6373 6190 97.13 97.50 97.31 e-vp 4768 4744 4649 98.00 97.50 97.75 i-vp 2602 2659 2526 95.00 97.08 96.03 e-adjp 384 369 322 87.26 83.85 85.52 i-pp 52 37 36 97.30 69.23 80.90 e-advp 822 812 708 87.19 86.13 86.66 i-advp 100 87 70 80.46 70.00 74.87 e-sbar 503 480 453 94.38 90.06 92.17 i-adjp 152 127 112 88.19 73.68 80.29 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 0 0 0.00 0.00 0.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.35 67.50 70.76 Avg2. 46451 46451 45185 97.27 97.27 97.27 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12265 11733 95.66 96.01 95.84 pp 4633 4674 4567 97.71 98.58 98.14 vp 4768 4744 4544 95.78 95.30 95.54 sbar 503 480 447 93.12 88.87 90.95 adjp 384 369 311 84.28 80.99 82.60 advp 822 812 701 86.33 85.28 85.80 prt 126 133 124 93.23 98.41 95.75 lst 10 0 0 0.00 0.00 0.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 74.61 69.34 71.88 Avg2. 23486 23485 22435 95.53 95.52 95.53 Current max chunk-based F1: 95.54 (iteration 41) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 43 Log-likelihood = -142970.895560 Norm (log-likelihood gradient vector) = 4366.522175 Norm (lambda vector) = 152.830760 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4678 4573 97.76 98.70 98.23 i-np 13660 13667 13365 97.79 97.84 97.82 e-np 12220 12253 12035 98.22 98.49 98.35 o 6349 6370 6190 97.17 97.50 97.33 e-vp 4768 4747 4654 98.04 97.61 97.82 i-vp 2602 2663 2529 94.97 97.19 96.07 e-adjp 384 367 323 88.01 84.11 86.02 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 811 710 87.55 86.37 86.96 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 476 450 94.54 89.46 91.93 i-adjp 152 126 112 88.89 73.68 80.58 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 3 3 100.00 30.00 46.15 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 79.49 68.82 73.77 Avg2. 46451 46451 45205 97.32 97.32 97.32 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12253 11739 95.81 96.06 95.93 pp 4633 4678 4566 97.61 98.55 98.08 vp 4768 4747 4546 95.77 95.34 95.55 sbar 503 476 444 93.28 88.27 90.70 adjp 384 367 311 84.74 80.99 82.82 advp 822 811 703 86.68 85.52 86.10 prt 126 133 124 93.23 98.41 95.75 lst 10 3 3 100.00 30.00 46.15 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 84.71 72.32 78.02 Avg2. 23486 23476 22444 95.60 95.56 95.58 Current max chunk-based F1: 95.58 (iteration 43) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 44 Log-likelihood = -138493.094107 Norm (log-likelihood gradient vector) = 4211.215475 Norm (lambda vector) = 155.767623 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4597 4530 98.54 97.78 98.16 i-np 13660 13453 13226 98.31 96.82 97.56 e-np 12220 12348 12061 97.68 98.70 98.18 o 6349 6495 6228 95.89 98.09 96.98 e-vp 4768 4748 4656 98.06 97.65 97.86 i-vp 2602 2667 2531 94.90 97.27 96.07 e-adjp 384 374 328 87.70 85.42 86.54 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 808 705 87.25 85.77 86.50 i-advp 100 83 68 81.93 68.00 74.32 e-sbar 503 545 476 87.34 94.63 90.84 i-adjp 152 122 107 87.70 70.39 78.10 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.90 69.30 73.35 Avg2. 46451 46451 45111 97.12 97.12 97.12 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12349 11748 95.13 96.14 95.63 pp 4633 4597 4523 98.39 97.63 98.01 vp 4768 4748 4547 95.77 95.36 95.57 sbar 503 545 470 86.24 93.44 89.69 adjp 384 374 312 83.42 81.25 82.32 advp 822 808 698 86.39 84.91 85.64 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.75 73.71 77.97 Avg2. 23486 23567 22434 95.19 95.52 95.36 Current max chunk-based F1: 95.58 (iteration 43) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 45 Log-likelihood = -141842.893982 Norm (log-likelihood gradient vector) = 20003.733764 Norm (lambda vector) = 161.627311 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4649 4560 98.09 98.42 98.25 i-np 13660 13549 13293 98.11 97.31 97.71 e-np 12220 12310 12049 97.88 98.60 98.24 o 6349 6432 6210 96.55 97.81 97.18 e-vp 4768 4747 4658 98.13 97.69 97.91 i-vp 2602 2665 2534 95.08 97.39 96.22 e-adjp 384 367 324 88.28 84.38 86.28 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 811 712 87.79 86.62 87.20 i-advp 100 84 68 80.95 68.00 73.91 e-sbar 503 501 462 92.22 91.85 92.03 i-adjp 152 126 112 88.89 73.68 80.58 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 3 3 100.00 30.00 46.15 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.61 69.28 73.65 Avg2. 46451 46451 45177 97.26 97.26 97.26 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12311 11747 95.42 96.13 95.77 pp 4633 4649 4553 97.94 98.27 98.10 vp 4768 4747 4552 95.89 95.47 95.68 sbar 503 501 457 91.22 90.85 91.04 adjp 384 367 311 84.74 80.99 82.82 advp 822 811 704 86.81 85.64 86.22 prt 126 133 124 93.23 98.41 95.75 lst 10 3 3 100.00 30.00 46.15 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.41 72.58 77.62 Avg2. 23486 23531 22459 95.44 95.63 95.54 Current max chunk-based F1: 95.58 (iteration 43) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 46 Log-likelihood = -136115.443808 Norm (log-likelihood gradient vector) = 8163.729388 Norm (lambda vector) = 158.066062 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4655 4566 98.09 98.55 98.32 i-np 13660 13674 13376 97.82 97.92 97.87 e-np 12220 12244 12029 98.24 98.44 98.34 o 6349 6378 6193 97.10 97.54 97.32 e-vp 4768 4746 4658 98.15 97.69 97.92 i-vp 2602 2673 2540 95.02 97.62 96.30 e-adjp 384 368 327 88.86 85.16 86.97 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 803 704 87.67 85.64 86.65 i-advp 100 81 68 83.95 68.00 75.14 e-sbar 503 494 460 93.12 91.45 92.28 i-adjp 152 124 111 89.52 73.03 80.43 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.89 69.74 74.04 Avg2. 46451 46451 45228 97.37 97.37 97.37 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12245 11740 95.88 96.07 95.97 pp 4633 4655 4559 97.94 98.40 98.17 vp 4768 4746 4552 95.91 95.47 95.69 sbar 503 494 455 92.11 90.46 91.27 adjp 384 368 313 85.05 81.51 83.24 advp 822 803 697 86.80 84.79 85.78 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.58 73.51 78.22 Avg2. 23486 23457 22452 95.72 95.60 95.66 Current max chunk-based F1: 95.66 (iteration 46) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 47 Log-likelihood = -132207.290369 Norm (log-likelihood gradient vector) = 3686.492862 Norm (lambda vector) = 159.675831 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4574 98.03 98.73 98.38 i-np 13660 13708 13399 97.75 98.09 97.92 e-np 12220 12223 12023 98.36 98.39 98.38 o 6349 6354 6184 97.32 97.40 97.36 e-vp 4768 4750 4659 98.08 97.71 97.90 i-vp 2602 2675 2540 94.95 97.62 96.27 e-adjp 384 373 329 88.20 85.68 86.92 i-pp 52 35 34 97.14 65.38 78.16 e-advp 822 801 704 87.89 85.64 86.75 i-advp 100 78 67 85.90 67.00 75.28 e-sbar 503 487 458 94.05 91.05 92.53 i-adjp 152 125 111 88.80 73.03 80.14 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.98 69.71 74.06 Avg2. 46451 46451 45244 97.40 97.40 97.40 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11732 95.98 96.01 95.99 pp 4633 4666 4567 97.88 98.58 98.23 vp 4768 4750 4551 95.81 95.45 95.63 sbar 503 487 453 93.02 90.06 91.52 adjp 384 373 315 84.45 82.03 83.22 advp 822 801 697 87.02 84.79 85.89 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.63 73.53 78.26 Avg2. 23486 23446 22451 95.76 95.59 95.67 Current max chunk-based F1: 95.67 (iteration 47) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 48 Log-likelihood = -129946.686068 Norm (log-likelihood gradient vector) = 3290.854445 Norm (lambda vector) = 159.917491 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4576 98.05 98.77 98.41 i-np 13660 13865 13460 97.08 98.54 97.80 e-np 12220 12142 11973 98.61 97.98 98.29 o 6349 6267 6133 97.86 96.60 97.23 e-vp 4768 4751 4660 98.08 97.73 97.91 i-vp 2602 2670 2538 95.06 97.54 96.28 e-adjp 384 375 329 87.73 85.68 86.69 i-pp 52 37 36 97.30 69.23 80.90 e-advp 822 811 708 87.30 86.13 86.71 i-advp 100 78 67 85.90 67.00 75.28 e-sbar 503 487 457 93.84 90.85 92.32 i-adjp 152 126 111 88.10 73.03 79.86 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.81 69.46 73.84 Avg2. 46451 46451 45209 97.33 97.33 97.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12143 11668 96.09 95.48 95.78 pp 4633 4667 4569 97.90 98.62 98.26 vp 4768 4751 4553 95.83 95.49 95.66 sbar 503 487 451 92.61 89.66 91.11 adjp 384 375 314 83.73 81.77 82.74 advp 822 811 701 86.44 85.28 85.85 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.47 73.47 78.15 Avg2. 23486 23380 22392 95.77 95.34 95.56 Current max chunk-based F1: 95.67 (iteration 47) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 49 Log-likelihood = -126995.484597 Norm (log-likelihood gradient vector) = 9873.406742 Norm (lambda vector) = 162.264757 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4677 4581 97.95 98.88 98.41 i-np 13660 13720 13401 97.67 98.10 97.89 e-np 12220 12214 12015 98.37 98.32 98.35 o 6349 6325 6168 97.52 97.15 97.33 e-vp 4768 4754 4662 98.06 97.78 97.92 i-vp 2602 2668 2537 95.09 97.50 96.28 e-adjp 384 378 329 87.04 85.68 86.35 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 812 709 87.32 86.25 86.78 i-advp 100 76 67 88.16 67.00 76.14 e-sbar 503 482 455 94.40 90.46 92.39 i-adjp 152 132 112 84.85 73.68 78.87 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.62 69.51 73.78 Avg2. 46451 46451 45233 97.38 97.38 97.38 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12214 11727 96.01 95.97 95.99 pp 4633 4677 4574 97.80 98.73 98.26 vp 4768 4754 4553 95.77 95.49 95.63 sbar 503 482 449 93.15 89.26 91.17 adjp 384 378 312 82.54 81.25 81.89 advp 822 812 703 86.58 85.52 86.05 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.40 73.46 78.12 Avg2. 23486 23463 22454 95.70 95.61 95.65 Current max chunk-based F1: 95.67 (iteration 47) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 50 Log-likelihood = -124942.424389 Norm (log-likelihood gradient vector) = 4392.344425 Norm (lambda vector) = 162.852111 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4679 4581 97.91 98.88 98.39 i-np 13660 13680 13374 97.76 97.91 97.83 e-np 12220 12231 12021 98.28 98.37 98.33 o 6349 6342 6174 97.35 97.24 97.30 e-vp 4768 4753 4660 98.04 97.73 97.89 i-vp 2602 2672 2537 94.95 97.50 96.21 e-adjp 384 380 330 86.84 85.94 86.39 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 809 706 87.27 85.89 86.57 i-advp 100 77 67 87.01 67.00 75.71 e-sbar 503 480 453 94.38 90.06 92.17 i-adjp 152 133 111 83.46 73.03 77.89 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 4 4 100.00 40.00 57.14 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.22 69.44 73.57 Avg2. 46451 46451 45211 97.33 97.33 97.33 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11731 95.91 96.00 95.96 pp 4633 4679 4572 97.71 98.68 98.20 vp 4768 4753 4551 95.75 95.45 95.60 sbar 503 480 447 93.12 88.87 90.95 adjp 384 380 312 82.11 81.25 81.68 advp 822 809 699 86.40 85.04 85.71 prt 126 133 124 93.23 98.41 95.75 lst 10 4 4 100.00 40.00 57.14 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.31 73.37 78.03 Avg2. 23486 23478 22448 95.61 95.58 95.60 Current max chunk-based F1: 95.67 (iteration 47) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 51 Log-likelihood = -123471.977574 Norm (log-likelihood gradient vector) = 3286.482690 Norm (lambda vector) = 163.712484 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4681 4582 97.89 98.90 98.39 i-np 13660 13619 13342 97.97 97.67 97.82 e-np 12220 12267 12042 98.17 98.54 98.35 o 6349 6367 6185 97.14 97.42 97.28 e-vp 4768 4746 4656 98.10 97.65 97.88 i-vp 2602 2674 2537 94.88 97.50 96.17 e-adjp 384 378 329 87.04 85.68 86.35 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 811 709 87.42 86.25 86.83 i-advp 100 79 68 86.08 68.00 75.98 e-sbar 503 481 456 94.80 90.66 92.68 i-adjp 152 132 111 84.09 73.03 78.17 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.24 70.03 73.91 Avg2. 46451 46451 45215 97.34 97.34 97.34 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12267 11753 95.81 96.18 95.99 pp 4633 4681 4573 97.69 98.70 98.20 vp 4768 4746 4548 95.83 95.39 95.61 sbar 503 481 450 93.56 89.46 91.46 adjp 384 378 312 82.54 81.25 81.89 advp 822 811 703 86.68 85.52 86.10 prt 126 133 124 93.23 98.41 95.75 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.42 74.49 78.70 Avg2. 23486 23511 22476 95.60 95.70 95.65 Current max chunk-based F1: 95.67 (iteration 47) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 52 Log-likelihood = -121436.580540 Norm (log-likelihood gradient vector) = 4049.882952 Norm (lambda vector) = 165.166342 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4680 4582 97.91 98.90 98.40 i-np 13660 13604 13342 98.07 97.67 97.87 e-np 12220 12278 12050 98.14 98.61 98.38 o 6349 6379 6196 97.13 97.59 97.36 e-vp 4768 4747 4656 98.08 97.65 97.87 i-vp 2602 2674 2537 94.88 97.50 96.17 e-adjp 384 376 330 87.77 85.94 86.84 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 803 705 87.80 85.77 86.77 i-advp 100 81 68 83.95 68.00 75.14 e-sbar 503 482 456 94.61 90.66 92.59 i-adjp 152 130 112 86.15 73.68 79.43 e-prt 126 134 124 92.54 98.41 95.38 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.25 70.06 73.93 Avg2. 46451 46451 45232 97.38 97.38 97.38 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12278 11764 95.81 96.27 96.04 pp 4633 4680 4572 97.69 98.68 98.19 vp 4768 4747 4548 95.81 95.39 95.60 sbar 503 482 450 93.36 89.46 91.37 adjp 384 376 316 84.04 82.29 83.16 advp 822 803 699 87.05 85.04 86.03 prt 126 134 124 92.54 98.41 95.38 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.52 74.55 78.78 Avg2. 23486 23514 22486 95.63 95.74 95.69 Current max chunk-based F1: 95.69 (iteration 52) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 53 Log-likelihood = -118954.132890 Norm (log-likelihood gradient vector) = 4454.707603 Norm (lambda vector) = 166.718434 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4574 98.01 98.73 98.37 i-np 13660 13580 13335 98.20 97.62 97.91 e-np 12220 12298 12062 98.08 98.71 98.39 o 6349 6399 6205 96.97 97.73 97.35 e-vp 4768 4749 4653 97.98 97.59 97.78 i-vp 2602 2666 2528 94.82 97.16 95.98 e-adjp 384 369 327 88.62 85.16 86.85 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 801 704 87.89 85.64 86.75 i-advp 100 83 68 81.93 68.00 74.32 e-sbar 503 491 460 93.69 91.45 92.56 i-adjp 152 131 112 85.50 73.68 79.15 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.15 70.26 74.00 Avg2. 46451 46451 45225 97.36 97.36 97.36 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12298 11781 95.80 96.41 96.10 pp 4633 4667 4564 97.79 98.51 98.15 vp 4768 4749 4543 95.66 95.28 95.47 sbar 503 491 455 92.67 90.46 91.55 adjp 384 369 313 84.82 81.51 83.13 advp 822 801 697 87.02 84.79 85.89 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.45 74.54 78.74 Avg2. 23486 23524 22490 95.60 95.76 95.68 Current max chunk-based F1: 95.69 (iteration 52) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 54 Log-likelihood = -114625.792689 Norm (log-likelihood gradient vector) = 5171.859211 Norm (lambda vector) = 170.182721 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4703 4585 97.49 98.96 98.22 i-np 13660 13838 13429 97.04 98.31 97.67 e-np 12220 12204 12000 98.33 98.20 98.26 o 6349 6286 6139 97.66 96.69 97.17 e-vp 4768 4730 4644 98.18 97.40 97.79 i-vp 2602 2669 2529 94.75 97.19 95.96 e-adjp 384 353 319 90.37 83.07 86.57 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 781 696 89.12 84.67 86.84 i-advp 100 87 70 80.46 70.00 74.87 e-sbar 503 454 436 96.04 86.68 91.12 i-adjp 152 126 112 88.89 73.68 80.58 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 12 10 83.33 83.33 83.33 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.49 70.03 74.02 Avg2. 46451 46451 45157 97.21 97.21 97.21 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12204 11686 95.76 95.63 95.69 pp 4633 4703 4573 97.24 98.70 97.96 vp 4768 4730 4531 95.79 95.03 95.41 sbar 503 454 433 95.37 86.08 90.49 adjp 384 353 307 86.97 79.95 83.31 advp 822 781 687 87.96 83.58 85.71 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.98 73.74 78.53 Avg2. 23486 23374 22354 95.64 95.18 95.41 Current max chunk-based F1: 95.69 (iteration 52) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 55 Log-likelihood = -115317.472412 Norm (log-likelihood gradient vector) = 11616.721631 Norm (lambda vector) = 174.563582 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4692 4584 97.70 98.94 98.32 i-np 13660 13730 13407 97.65 98.15 97.90 e-np 12220 12240 12036 98.33 98.49 98.41 o 6349 6331 6172 97.49 97.21 97.35 e-vp 4768 4740 4649 98.08 97.50 97.79 i-vp 2602 2668 2528 94.75 97.16 95.94 e-adjp 384 360 323 89.72 84.11 86.83 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 796 703 88.32 85.52 86.90 i-advp 100 83 68 81.93 68.00 74.32 e-sbar 503 466 446 95.71 88.67 92.05 i-adjp 152 127 112 88.19 73.68 80.29 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.34 70.06 73.97 Avg2. 46451 46451 45225 97.36 97.36 97.36 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11756 96.05 96.20 96.12 pp 4633 4692 4573 97.46 98.70 98.08 vp 4768 4740 4538 95.74 95.18 95.46 sbar 503 466 442 94.85 87.87 91.23 adjp 384 360 309 85.83 80.47 83.06 advp 822 796 696 87.44 84.67 86.03 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.81 74.15 78.69 Avg2. 23486 23443 22451 95.77 95.59 95.68 Current max chunk-based F1: 95.69 (iteration 52) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 56 Log-likelihood = -112491.476015 Norm (log-likelihood gradient vector) = 5610.892515 Norm (lambda vector) = 172.217901 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4674 4577 97.92 98.79 98.36 i-np 13660 13693 13408 97.92 98.16 98.04 e-np 12220 12256 12054 98.35 98.64 98.50 o 6349 6359 6194 97.41 97.56 97.48 e-vp 4768 4747 4655 98.06 97.63 97.85 i-vp 2602 2654 2523 95.06 96.96 96.00 e-adjp 384 359 324 90.25 84.38 87.21 i-pp 52 40 34 85.00 65.38 73.91 e-advp 822 797 705 88.46 85.77 87.09 i-advp 100 83 68 81.93 68.00 74.32 e-sbar 503 481 455 94.59 90.46 92.48 i-adjp 152 129 114 88.37 75.00 81.14 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.26 70.26 74.04 Avg2. 46451 46451 45274 97.47 97.47 97.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12256 11794 96.23 96.51 96.37 pp 4633 4674 4565 97.67 98.53 98.10 vp 4768 4747 4545 95.74 95.32 95.53 sbar 503 481 451 93.76 89.66 91.67 adjp 384 359 311 86.63 80.99 83.71 advp 822 797 698 87.58 84.91 86.23 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.84 74.43 78.86 Avg2. 23486 23463 22501 95.90 95.81 95.85 Current max chunk-based F1: 95.85 (iteration 56) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 57 Log-likelihood = -109943.281320 Norm (log-likelihood gradient vector) = 3032.412179 Norm (lambda vector) = 173.810834 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4573 98.03 98.70 98.37 i-np 13660 13733 13423 97.74 98.27 98.00 e-np 12220 12225 12036 98.45 98.49 98.47 o 6349 6341 6183 97.51 97.39 97.45 e-vp 4768 4743 4656 98.17 97.65 97.91 i-vp 2602 2656 2530 95.26 97.23 96.23 e-adjp 384 364 326 89.56 84.90 87.17 i-pp 52 40 34 85.00 65.38 73.91 e-advp 822 800 709 88.62 86.25 87.42 i-advp 100 80 69 86.25 69.00 76.67 e-sbar 503 490 460 93.88 91.45 92.65 i-adjp 152 133 114 85.71 75.00 80.00 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 19 18 94.74 75.00 83.72 e-conjp 16 10 9 90.00 56.25 69.23 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.29 70.72 74.31 Avg2. 46451 46451 45278 97.47 97.47 97.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12225 11773 96.30 96.34 96.32 pp 4633 4665 4561 97.77 98.45 98.11 vp 4768 4743 4551 95.95 95.45 95.70 sbar 503 490 455 92.86 90.46 91.64 adjp 384 364 312 85.71 81.25 83.42 advp 822 800 703 87.88 85.52 86.68 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.83 75.21 79.29 Avg2. 23486 23437 22493 95.97 95.77 95.87 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 58 Log-likelihood = -107973.260160 Norm (log-likelihood gradient vector) = 3316.607229 Norm (lambda vector) = 175.040262 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4570 97.98 98.64 98.31 i-np 13660 13584 13342 98.22 97.67 97.94 e-np 12220 12285 12059 98.16 98.68 98.42 o 6349 6409 6214 96.96 97.87 97.41 e-vp 4768 4749 4659 98.10 97.71 97.91 i-vp 2602 2661 2533 95.19 97.35 96.26 e-adjp 384 369 327 88.62 85.16 86.85 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 804 709 88.18 86.25 87.21 i-advp 100 81 68 83.95 68.00 75.14 e-sbar 503 491 460 93.69 91.45 92.56 i-adjp 152 136 113 83.09 74.34 78.47 e-prt 126 136 124 91.18 98.41 94.66 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.93 69.93 73.71 Avg2. 46451 46451 45250 97.41 97.41 97.41 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12285 11785 95.93 96.44 96.18 pp 4633 4664 4559 97.75 98.40 98.07 vp 4768 4749 4553 95.87 95.49 95.68 sbar 503 491 455 92.67 90.46 91.55 adjp 384 369 311 84.28 80.99 82.60 advp 822 804 702 87.31 85.40 86.35 prt 126 136 124 91.18 98.41 94.66 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.39 74.56 78.73 Avg2. 23486 23512 22502 95.70 95.81 95.76 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 59 Log-likelihood = -106720.887408 Norm (log-likelihood gradient vector) = 5709.084806 Norm (lambda vector) = 175.950012 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4569 97.98 98.62 98.30 i-np 13660 13730 13421 97.75 98.25 98.00 e-np 12220 12213 12029 98.49 98.44 98.47 o 6349 6339 6181 97.51 97.35 97.43 e-vp 4768 4742 4656 98.19 97.65 97.92 i-vp 2602 2665 2536 95.16 97.46 96.30 e-adjp 384 370 328 88.65 85.42 87.00 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 804 711 88.43 86.50 87.45 i-advp 100 81 68 83.95 68.00 75.14 e-sbar 503 490 458 93.47 91.05 92.25 i-adjp 152 136 113 83.09 74.34 78.47 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 5 5 100.00 50.00 66.67 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.00 70.02 73.80 Avg2. 46451 46451 45267 97.45 97.45 97.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11756 96.26 96.20 96.23 pp 4633 4663 4558 97.75 98.38 98.06 vp 4768 4742 4550 95.95 95.43 95.69 sbar 503 490 453 92.45 90.06 91.24 adjp 384 370 312 84.32 81.25 82.76 advp 822 804 704 87.56 85.64 86.59 prt 126 135 124 91.85 98.41 95.02 lst 10 5 5 100.00 50.00 66.67 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.50 74.54 78.77 Avg2. 23486 23431 22470 95.90 95.67 95.79 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 60 Log-likelihood = -104723.294632 Norm (log-likelihood gradient vector) = 3724.828238 Norm (lambda vector) = 176.569476 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4575 98.01 98.75 98.38 i-np 13660 13763 13440 97.65 98.39 98.02 e-np 12220 12198 12024 98.57 98.40 98.48 o 6349 6314 6169 97.70 97.16 97.43 e-vp 4768 4742 4655 98.17 97.63 97.90 i-vp 2602 2668 2536 95.05 97.46 96.24 e-adjp 384 370 329 88.92 85.68 87.27 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 805 711 88.32 86.50 87.40 i-advp 100 81 68 83.95 68.00 75.14 e-sbar 503 486 458 94.24 91.05 92.62 i-adjp 152 135 113 83.70 74.34 78.75 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 8 7 87.50 70.00 77.78 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.46 71.04 74.11 Avg2. 46451 46451 45277 97.47 97.47 97.47 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12198 11752 96.34 96.17 96.26 pp 4633 4668 4564 97.77 98.51 98.14 vp 4768 4742 4547 95.89 95.36 95.63 sbar 503 486 453 93.21 90.06 91.61 adjp 384 370 311 84.05 80.99 82.49 advp 822 805 704 87.45 85.64 86.54 prt 126 135 124 91.85 98.41 95.02 lst 10 8 7 87.50 70.00 77.78 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.30 76.52 79.30 Avg2. 23486 23421 22470 95.94 95.67 95.81 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 61 Log-likelihood = -102952.491558 Norm (log-likelihood gradient vector) = 3790.455089 Norm (lambda vector) = 177.242083 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4575 97.92 98.75 98.33 i-np 13660 13745 13432 97.72 98.33 98.03 e-np 12220 12210 12031 98.53 98.45 98.49 o 6349 6310 6168 97.75 97.15 97.45 e-vp 4768 4737 4656 98.29 97.65 97.97 i-vp 2602 2676 2543 95.03 97.73 96.36 e-adjp 384 372 330 88.71 85.94 87.30 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 804 712 88.56 86.62 87.58 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 483 456 94.41 90.66 92.49 i-adjp 152 136 114 83.82 75.00 79.17 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 8 7 87.50 70.00 77.78 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.04 71.13 73.97 Avg2. 46451 46451 45285 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12210 11760 96.31 96.24 96.28 pp 4633 4672 4564 97.69 98.51 98.10 vp 4768 4737 4551 96.07 95.45 95.76 sbar 503 483 450 93.17 89.46 91.28 adjp 384 372 313 84.14 81.51 82.80 advp 822 804 704 87.56 85.64 86.59 prt 126 135 124 91.85 98.41 95.02 lst 10 8 7 87.50 70.00 77.78 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.32 76.52 79.31 Avg2. 23486 23430 22481 95.95 95.72 95.84 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 62 Log-likelihood = -99989.545753 Norm (log-likelihood gradient vector) = 2721.856351 Norm (lambda vector) = 179.495459 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4573 97.88 98.70 98.29 i-np 13660 13697 13392 97.77 98.04 97.91 e-np 12220 12239 12036 98.34 98.49 98.42 o 6349 6340 6179 97.46 97.32 97.39 e-vp 4768 4737 4659 98.35 97.71 98.03 i-vp 2602 2676 2546 95.14 97.85 96.48 e-adjp 384 370 328 88.65 85.42 87.00 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 794 708 89.17 86.13 87.62 i-advp 100 82 69 84.15 69.00 75.82 e-sbar 503 481 453 94.18 90.06 92.07 i-adjp 152 138 116 84.06 76.32 80.00 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 18 17 94.44 70.83 80.95 e-conjp 16 10 9 90.00 56.25 69.23 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 9 7 77.78 70.00 73.68 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.89 71.96 74.35 Avg2. 46451 46451 45260 97.44 97.44 97.44 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12239 11755 96.05 96.19 96.12 pp 4633 4672 4562 97.65 98.47 98.05 vp 4768 4737 4556 96.18 95.55 95.87 sbar 503 481 448 93.14 89.07 91.06 adjp 384 370 313 84.59 81.51 83.02 advp 822 794 700 88.16 85.16 86.63 prt 126 135 124 91.85 98.41 95.02 lst 10 9 7 77.78 70.00 73.68 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.54 77.06 79.24 Avg2. 23486 23447 22474 95.85 95.69 95.77 Current max chunk-based F1: 95.87 (iteration 57) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 63 Log-likelihood = -96913.952167 Norm (log-likelihood gradient vector) = 3779.272196 Norm (lambda vector) = 182.557545 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4671 4574 97.92 98.73 98.32 i-np 13660 13718 13417 97.81 98.22 98.01 e-np 12220 12224 12034 98.45 98.48 98.46 o 6349 6330 6180 97.63 97.34 97.48 e-vp 4768 4736 4656 98.31 97.65 97.98 i-vp 2602 2681 2546 94.96 97.85 96.38 e-adjp 384 369 328 88.89 85.42 87.12 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 798 712 89.22 86.62 87.90 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 483 455 94.20 90.46 92.29 i-adjp 152 134 115 85.82 75.66 80.42 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 17 16 94.12 66.67 78.05 e-conjp 16 9 8 88.89 50.00 64.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.20 72.38 74.71 Avg2. 46451 46451 45287 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11769 96.28 96.31 96.29 pp 4633 4671 4564 97.71 98.51 98.11 vp 4768 4736 4552 96.11 95.47 95.79 sbar 503 483 450 93.17 89.46 91.28 adjp 384 369 316 85.64 82.29 83.93 advp 822 798 704 88.22 85.64 86.91 prt 126 135 124 91.85 98.41 95.02 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 8 88.89 50.00 64.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.79 77.61 79.64 Avg2. 23486 23435 22495 95.99 95.78 95.88 Current max chunk-based F1: 95.88 (iteration 63) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 64 Log-likelihood = -94732.803212 Norm (log-likelihood gradient vector) = 2483.136525 Norm (lambda vector) = 184.611897 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4573 98.03 98.70 98.37 i-np 13660 13694 13416 97.97 98.21 98.09 e-np 12220 12231 12044 98.47 98.56 98.52 o 6349 6348 6195 97.59 97.57 97.58 e-vp 4768 4738 4659 98.33 97.71 98.02 i-vp 2602 2679 2541 94.85 97.66 96.23 e-adjp 384 368 328 89.13 85.42 87.23 i-pp 52 38 34 89.47 65.38 75.56 e-advp 822 796 709 89.07 86.25 87.64 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 491 460 93.69 91.45 92.56 i-adjp 152 131 114 87.02 75.00 80.57 e-prt 126 135 124 91.85 98.41 95.02 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 19 18 94.74 75.00 83.72 e-conjp 16 10 9 90.00 56.25 69.23 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.33 73.11 75.16 Avg2. 46451 46451 45312 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11787 96.37 96.46 96.41 pp 4633 4665 4563 97.81 98.49 98.15 vp 4768 4738 4550 96.03 95.43 95.73 sbar 503 491 455 92.67 90.46 91.55 adjp 384 368 316 85.87 82.29 84.04 advp 822 796 700 87.94 85.16 86.53 prt 126 135 124 91.85 98.41 95.02 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 9 90.00 56.25 69.23 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.85 78.29 80.03 Avg2. 23486 23444 22512 96.02 95.85 95.94 Current max chunk-based F1: 95.94 (iteration 64) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 65 Log-likelihood = -93034.073359 Norm (log-likelihood gradient vector) = 2668.729186 Norm (lambda vector) = 186.366514 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4578 97.99 98.81 98.40 i-np 13660 13728 13407 97.66 98.15 97.90 e-np 12220 12217 12027 98.44 98.42 98.43 o 6349 6335 6177 97.51 97.29 97.40 e-vp 4768 4735 4660 98.42 97.73 98.07 i-vp 2602 2669 2537 95.05 97.50 96.26 e-adjp 384 366 327 89.34 85.16 87.20 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 802 709 88.40 86.25 87.32 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 484 460 95.04 91.45 93.21 i-adjp 152 131 114 87.02 75.00 80.57 e-prt 126 134 123 91.79 97.62 94.62 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 20 19 95.00 79.17 86.36 e-conjp 16 11 10 90.91 62.50 74.07 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.31 73.56 75.39 Avg2. 46451 46451 45270 97.46 97.46 97.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11755 96.22 96.19 96.21 pp 4633 4672 4567 97.75 98.58 98.16 vp 4768 4735 4550 96.09 95.43 95.76 sbar 503 484 455 94.01 90.46 92.20 adjp 384 366 315 86.07 82.03 84.00 advp 822 802 700 87.28 85.16 86.21 prt 126 134 123 91.79 97.62 94.62 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 10 90.91 62.50 74.07 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.01 78.80 80.37 Avg2. 23486 23431 22483 95.95 95.73 95.84 Current max chunk-based F1: 95.94 (iteration 64) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 66 Log-likelihood = -90723.476884 Norm (log-likelihood gradient vector) = 3241.036472 Norm (lambda vector) = 189.601010 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4658 4569 98.09 98.62 98.35 i-np 13660 13658 13390 98.04 98.02 98.03 e-np 12220 12245 12043 98.35 98.55 98.45 o 6349 6369 6196 97.28 97.59 97.44 e-vp 4768 4736 4659 98.37 97.71 98.04 i-vp 2602 2670 2536 94.98 97.46 96.21 e-adjp 384 372 332 89.25 86.46 87.83 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 808 714 88.37 86.86 87.61 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 495 462 93.33 91.85 92.59 i-adjp 152 133 115 86.47 75.66 80.70 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 17 17 100.00 70.83 82.93 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.96 73.04 75.42 Avg2. 46451 46451 45288 97.50 97.50 97.50 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12245 11783 96.23 96.42 96.33 pp 4633 4658 4559 97.87 98.40 98.14 vp 4768 4736 4547 96.01 95.36 95.69 sbar 503 495 457 92.32 90.85 91.58 adjp 384 372 319 85.75 83.07 84.39 advp 822 808 707 87.50 86.01 86.75 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.74 78.32 80.47 Avg2. 23486 23466 22511 95.93 95.85 95.89 Current max chunk-based F1: 95.94 (iteration 64) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 67 Log-likelihood = -89117.361153 Norm (log-likelihood gradient vector) = 3395.885436 Norm (lambda vector) = 192.232819 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4660 4573 98.13 98.70 98.42 i-np 13660 13776 13435 97.52 98.35 97.94 e-np 12220 12187 12011 98.56 98.29 98.42 o 6349 6306 6163 97.73 97.07 97.40 e-vp 4768 4735 4660 98.42 97.73 98.07 i-vp 2602 2671 2539 95.06 97.58 96.30 e-adjp 384 373 333 89.28 86.72 87.98 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 491 462 94.09 91.85 92.96 i-adjp 152 133 115 86.47 75.66 80.70 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 17 17 100.00 70.83 82.93 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.03 73.07 75.47 Avg2. 46451 46451 45282 97.48 97.48 97.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12187 11742 96.35 96.09 96.22 pp 4633 4660 4563 97.92 98.49 98.20 vp 4768 4735 4550 96.09 95.43 95.76 sbar 503 491 457 93.08 90.85 91.95 adjp 384 373 320 85.79 83.33 84.54 advp 822 812 712 87.68 86.62 87.15 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.86 78.39 80.56 Avg2. 23486 23410 22483 96.04 95.73 95.88 Current max chunk-based F1: 95.94 (iteration 64) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 68 Log-likelihood = -87776.303767 Norm (log-likelihood gradient vector) = 3096.446671 Norm (lambda vector) = 193.449991 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4576 97.95 98.77 98.36 i-np 13660 13717 13421 97.84 98.25 98.05 e-np 12220 12220 12032 98.46 98.46 98.46 o 6349 6333 6181 97.60 97.35 97.48 e-vp 4768 4734 4659 98.42 97.71 98.06 i-vp 2602 2671 2539 95.06 97.58 96.30 e-adjp 384 375 333 88.80 86.72 87.75 i-pp 52 40 34 85.00 65.38 73.91 e-advp 822 814 720 88.45 87.59 88.02 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 481 458 95.22 91.05 93.09 i-adjp 152 137 116 84.67 76.32 80.28 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 11 11 100.00 45.83 62.86 e-conjp 16 6 6 100.00 37.50 54.55 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.97 70.43 74.01 Avg2. 46451 46451 45296 97.51 97.51 97.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12220 11773 96.34 96.34 96.34 pp 4633 4672 4567 97.75 98.58 98.16 vp 4768 4734 4551 96.13 95.45 95.79 sbar 503 481 454 94.39 90.26 92.28 adjp 384 375 319 85.07 83.07 84.06 advp 822 814 714 87.71 86.86 87.29 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 6 6 100.00 37.50 54.55 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.91 76.49 79.57 Avg2. 23486 23445 22514 96.03 95.86 95.95 Current max chunk-based F1: 95.95 (iteration 68) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 69 Log-likelihood = -86367.055417 Norm (log-likelihood gradient vector) = 2495.008893 Norm (lambda vector) = 194.744539 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4577 97.97 98.79 98.38 i-np 13660 13780 13442 97.55 98.40 97.97 e-np 12220 12191 12012 98.53 98.30 98.41 o 6349 6303 6161 97.75 97.04 97.39 e-vp 4768 4731 4660 98.50 97.73 98.12 i-vp 2602 2669 2541 95.20 97.66 96.41 e-adjp 384 375 333 88.80 86.72 87.75 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 813 721 88.68 87.71 88.20 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 482 459 95.23 91.25 93.20 i-adjp 152 136 116 85.29 76.32 80.56 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 13 13 100.00 54.17 70.27 e-conjp 16 7 7 100.00 43.75 60.87 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.13 71.17 74.49 Avg2. 46451 46451 45286 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12191 11741 96.31 96.08 96.19 pp 4633 4672 4568 97.77 98.60 98.18 vp 4768 4731 4556 96.30 95.55 95.93 sbar 503 482 455 94.40 90.46 92.39 adjp 384 375 320 85.33 83.33 84.32 advp 822 813 715 87.95 86.98 87.46 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 7 7 100.00 43.75 60.87 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.98 77.16 79.96 Avg2. 23486 23414 22492 96.06 95.77 95.91 Current max chunk-based F1: 95.95 (iteration 68) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 70 Log-likelihood = -84795.935765 Norm (log-likelihood gradient vector) = 3806.821307 Norm (lambda vector) = 196.254169 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4579 98.01 98.83 98.42 i-np 13660 13713 13410 97.79 98.17 97.98 e-np 12220 12228 12031 98.39 98.45 98.42 o 6349 6335 6178 97.52 97.31 97.41 e-vp 4768 4730 4662 98.56 97.78 98.17 i-vp 2602 2668 2543 95.31 97.73 96.51 e-adjp 384 377 334 88.59 86.98 87.78 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 805 720 89.44 87.59 88.51 i-advp 100 81 70 86.42 70.00 77.35 e-sbar 503 482 459 95.23 91.25 93.20 i-adjp 152 137 118 86.13 77.63 81.66 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 17 100.00 70.83 82.93 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.27 72.81 75.44 Avg2. 46451 46451 45305 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12228 11758 96.16 96.22 96.19 pp 4633 4672 4570 97.82 98.64 98.23 vp 4768 4730 4558 96.36 95.60 95.98 sbar 503 482 455 94.40 90.46 92.39 adjp 384 377 321 85.15 83.59 84.36 advp 822 805 714 88.70 86.86 87.77 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.03 78.44 80.67 Avg2. 23486 23446 22515 96.03 95.87 95.95 Current max chunk-based F1: 95.95 (iteration 70) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 71 Log-likelihood = -82769.781144 Norm (log-likelihood gradient vector) = 2268.237281 Norm (lambda vector) = 198.664004 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4576 97.99 98.77 98.38 i-np 13660 13713 13418 97.85 98.23 98.04 e-np 12220 12227 12033 98.41 98.47 98.44 o 6349 6332 6179 97.58 97.32 97.45 e-vp 4768 4732 4663 98.54 97.80 98.17 i-vp 2602 2667 2543 95.35 97.73 96.53 e-adjp 384 376 334 88.83 86.98 87.89 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 806 720 89.33 87.59 88.45 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 487 460 94.46 91.45 92.93 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 17 17 100.00 70.83 82.93 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.06 72.81 75.34 Avg2. 46451 46451 45314 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11767 96.24 96.29 96.27 pp 4633 4670 4566 97.77 98.55 98.16 vp 4768 4732 4559 96.34 95.62 95.98 sbar 503 487 456 93.63 90.66 92.12 adjp 384 376 320 85.11 83.33 84.21 advp 822 806 713 88.46 86.74 87.59 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.93 78.43 80.61 Avg2. 23486 23450 22520 96.03 95.89 95.96 Current max chunk-based F1: 95.96 (iteration 71) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 72 Log-likelihood = -81568.996194 Norm (log-likelihood gradient vector) = 2138.383202 Norm (lambda vector) = 200.387858 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4675 4579 97.95 98.83 98.39 i-np 13660 13666 13373 97.86 97.90 97.88 e-np 12220 12257 12036 98.20 98.49 98.35 o 6349 6365 6183 97.14 97.39 97.26 e-vp 4768 4733 4662 98.50 97.78 98.14 i-vp 2602 2659 2537 95.41 97.50 96.45 e-adjp 384 373 332 89.01 86.46 87.71 i-pp 52 39 34 87.18 65.38 74.73 e-advp 822 804 718 89.30 87.35 88.31 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 479 456 95.20 90.66 92.87 i-adjp 152 131 115 87.79 75.66 81.27 e-prt 126 133 122 91.73 96.83 94.21 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 19 19 100.00 79.17 88.37 e-conjp 16 10 10 100.00 62.50 76.92 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.29 73.31 75.72 Avg2. 46451 46451 45265 97.45 97.45 97.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12257 11753 95.89 96.18 96.03 pp 4633 4675 4570 97.75 98.64 98.20 vp 4768 4733 4557 96.28 95.57 95.93 sbar 503 479 452 94.36 89.86 92.06 adjp 384 373 317 84.99 82.55 83.75 advp 822 804 711 88.43 86.50 87.45 prt 126 133 122 91.73 96.83 94.21 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 10 100.00 62.50 76.92 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.94 78.86 80.85 Avg2. 23486 23474 22500 95.85 95.80 95.83 Current max chunk-based F1: 95.96 (iteration 71) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 73 Log-likelihood = -80177.897742 Norm (log-likelihood gradient vector) = 4000.101498 Norm (lambda vector) = 203.560997 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4576 98.01 98.77 98.39 i-np 13660 13693 13417 97.98 98.22 98.10 e-np 12220 12245 12044 98.36 98.56 98.46 o 6349 6352 6191 97.47 97.51 97.49 e-vp 4768 4734 4663 98.50 97.80 98.15 i-vp 2602 2661 2539 95.42 97.58 96.48 e-adjp 384 372 332 89.25 86.46 87.83 i-pp 52 37 34 91.89 65.38 76.40 e-advp 822 805 717 89.07 87.23 88.14 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 483 458 94.82 91.05 92.90 i-adjp 152 130 115 88.46 75.66 81.56 e-prt 126 132 122 92.42 96.83 94.57 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 19 19 100.00 79.17 88.37 e-conjp 16 10 10 100.00 62.50 76.92 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.33 73.35 75.76 Avg2. 46451 46451 45326 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12245 11789 96.28 96.47 96.37 pp 4633 4669 4568 97.84 98.60 98.22 vp 4768 4734 4560 96.32 95.64 95.98 sbar 503 483 453 93.79 90.06 91.89 adjp 384 372 317 85.22 82.55 83.86 advp 822 805 710 88.20 86.37 87.28 prt 126 132 122 92.42 96.83 94.57 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 10 100.00 62.50 76.92 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.01 78.90 80.90 Avg2. 23486 23460 22537 96.07 95.96 96.01 Current max chunk-based F1: 96.01 (iteration 73) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 74 Log-likelihood = -78882.001837 Norm (log-likelihood gradient vector) = 2132.300568 Norm (lambda vector) = 205.706529 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4574 98.11 98.73 98.42 i-np 13660 13706 13417 97.89 98.22 98.06 e-np 12220 12228 12038 98.45 98.51 98.48 o 6349 6341 6186 97.56 97.43 97.49 e-vp 4768 4735 4663 98.48 97.80 98.14 i-vp 2602 2660 2537 95.38 97.50 96.43 e-adjp 384 377 336 89.12 87.50 88.30 i-pp 52 37 34 91.89 65.38 76.40 e-advp 822 812 720 88.67 87.59 88.13 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 488 461 94.47 91.65 93.04 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 132 122 92.42 96.83 94.57 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 19 19 100.00 79.17 88.37 e-conjp 16 10 10 100.00 62.50 76.92 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.15 73.47 75.74 Avg2. 46451 46451 45322 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12228 11780 96.34 96.40 96.37 pp 4633 4662 4566 97.94 98.55 98.25 vp 4768 4735 4558 96.26 95.60 95.93 sbar 503 488 456 93.44 90.66 92.03 adjp 384 377 321 85.15 83.59 84.36 advp 822 812 713 87.81 86.74 87.27 prt 126 132 122 92.42 96.83 94.57 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 10 100.00 62.50 76.92 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.94 79.09 80.97 Avg2. 23486 23454 22534 96.08 95.95 96.01 Current max chunk-based F1: 96.01 (iteration 73) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 75 Log-likelihood = -78297.884780 Norm (log-likelihood gradient vector) = 1907.098011 Norm (lambda vector) = 206.721936 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4577 98.11 98.79 98.45 i-np 13660 13775 13430 97.50 98.32 97.90 e-np 12220 12193 12011 98.51 98.29 98.40 o 6349 6306 6159 97.67 97.01 97.34 e-vp 4768 4733 4663 98.52 97.80 98.16 i-vp 2602 2663 2537 95.27 97.50 96.37 e-adjp 384 374 335 89.57 87.24 88.39 i-pp 52 37 34 91.89 65.38 76.40 e-advp 822 811 720 88.78 87.59 88.18 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 484 459 94.83 91.25 93.01 i-adjp 152 139 117 84.17 76.97 80.41 e-prt 126 132 122 92.42 96.83 94.57 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 19 19 100.00 79.17 88.37 e-conjp 16 10 10 100.00 62.50 76.92 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.06 73.45 75.68 Avg2. 46451 46451 45282 97.48 97.48 97.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11743 96.31 96.10 96.20 pp 4633 4665 4569 97.94 98.62 98.28 vp 4768 4733 4556 96.26 95.55 95.91 sbar 503 484 454 93.80 90.26 92.00 adjp 384 374 320 85.56 83.33 84.43 advp 822 811 713 87.92 86.74 87.32 prt 126 132 122 92.42 96.83 94.57 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 10 100.00 62.50 76.92 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.02 78.99 80.96 Avg2. 23486 23412 22495 96.08 95.78 95.93 Current max chunk-based F1: 96.01 (iteration 73) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 76 Log-likelihood = -77398.700287 Norm (log-likelihood gradient vector) = 2755.399659 Norm (lambda vector) = 207.992487 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4655 4574 98.26 98.73 98.49 i-np 13660 13669 13410 98.11 98.17 98.14 e-np 12220 12239 12047 98.43 98.58 98.51 o 6349 6365 6200 97.41 97.65 97.53 e-vp 4768 4736 4667 98.54 97.88 98.21 i-vp 2602 2662 2540 95.42 97.62 96.50 e-adjp 384 374 336 89.84 87.50 88.65 i-pp 52 36 34 94.44 65.38 77.27 e-advp 822 809 718 88.75 87.35 88.04 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 495 465 93.94 92.45 93.19 i-adjp 152 145 123 84.83 80.92 82.83 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 15 15 100.00 62.50 76.92 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.13 72.70 75.32 Avg2. 46451 46451 45349 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12239 11799 96.40 96.55 96.48 pp 4633 4655 4566 98.09 98.55 98.32 vp 4768 4736 4564 96.37 95.72 96.04 sbar 503 495 460 92.93 91.45 92.18 adjp 384 374 323 86.36 84.11 85.22 advp 822 809 712 88.01 86.62 87.31 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.06 78.06 80.49 Avg2. 23486 23459 22563 96.18 96.07 96.13 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 77 Log-likelihood = -76294.847262 Norm (log-likelihood gradient vector) = 2770.679459 Norm (lambda vector) = 209.877474 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4577 98.20 98.79 98.49 i-np 13660 13764 13446 97.69 98.43 98.06 e-np 12220 12195 12017 98.54 98.34 98.44 o 6349 6323 6173 97.63 97.23 97.43 e-vp 4768 4735 4665 98.52 97.84 98.18 i-vp 2602 2660 2536 95.34 97.46 96.39 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 37 35 94.59 67.31 78.65 e-advp 822 804 718 89.30 87.35 88.31 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 492 463 94.11 92.05 93.07 i-adjp 152 146 123 84.25 80.92 82.55 e-prt 126 131 122 93.13 96.83 94.94 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 15 15 100.00 62.50 76.92 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.26 72.68 75.37 Avg2. 46451 46451 45321 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12195 11757 96.41 96.21 96.31 pp 4633 4661 4571 98.07 98.66 98.36 vp 4768 4735 4562 96.35 95.68 96.01 sbar 503 492 458 93.09 91.05 92.06 adjp 384 372 321 86.29 83.59 84.92 advp 822 804 712 88.56 86.62 87.58 prt 126 131 122 93.13 96.83 94.94 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.19 77.86 80.44 Avg2. 23486 23408 22519 96.20 95.88 96.04 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 78 Log-likelihood = -75027.302701 Norm (log-likelihood gradient vector) = 4356.283956 Norm (lambda vector) = 211.795192 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4579 98.05 98.83 98.44 i-np 13660 13707 13426 97.95 98.29 98.12 e-np 12220 12219 12033 98.48 98.47 98.47 o 6349 6350 6191 97.50 97.51 97.50 e-vp 4768 4735 4666 98.54 97.86 98.20 i-vp 2602 2664 2540 95.35 97.62 96.47 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 37 35 94.59 67.31 78.65 e-advp 822 804 717 89.18 87.23 88.19 i-advp 100 81 70 86.42 70.00 77.35 e-sbar 503 488 461 94.47 91.65 93.04 i-adjp 152 144 123 85.42 80.92 83.11 e-prt 126 131 122 93.13 96.83 94.94 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 15 15 100.00 62.50 76.92 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.38 72.68 75.42 Avg2. 46451 46451 45339 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11783 96.43 96.42 96.43 pp 4633 4670 4573 97.92 98.70 98.31 vp 4768 4735 4563 96.37 95.70 96.03 sbar 503 488 456 93.44 90.66 92.03 adjp 384 372 321 86.29 83.59 84.92 advp 822 804 711 88.43 86.50 87.45 prt 126 131 122 93.13 96.83 94.94 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.20 77.84 80.43 Avg2. 23486 23437 22545 96.19 95.99 96.09 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 79 Log-likelihood = -73732.567947 Norm (log-likelihood gradient vector) = 2087.875427 Norm (lambda vector) = 213.436188 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4578 98.03 98.81 98.42 i-np 13660 13692 13414 97.97 98.20 98.08 e-np 12220 12227 12033 98.41 98.47 98.44 o 6349 6359 6194 97.41 97.56 97.48 e-vp 4768 4737 4667 98.52 97.88 98.20 i-vp 2602 2663 2540 95.38 97.62 96.49 e-adjp 384 371 333 89.76 86.72 88.21 i-pp 52 38 35 92.11 67.31 77.78 e-advp 822 804 716 89.05 87.10 88.07 i-advp 100 81 70 86.42 70.00 77.35 e-sbar 503 488 461 94.47 91.65 93.04 i-adjp 152 142 121 85.21 79.61 82.31 e-prt 126 130 121 93.08 96.03 94.53 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 15 15 100.00 62.50 76.92 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.23 72.56 75.29 Avg2. 46451 46451 45325 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11781 96.35 96.41 96.38 pp 4633 4670 4572 97.90 98.68 98.29 vp 4768 4737 4564 96.35 95.72 96.03 sbar 503 488 456 93.44 90.66 92.03 adjp 384 371 319 85.98 83.07 84.50 advp 822 804 710 88.31 86.37 87.33 prt 126 130 121 93.08 96.03 94.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.14 77.69 80.33 Avg2. 23486 23445 22539 96.14 95.97 96.05 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 80 Log-likelihood = -73054.873100 Norm (log-likelihood gradient vector) = 1897.975654 Norm (lambda vector) = 214.033909 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4579 98.01 98.83 98.42 i-np 13660 13660 13399 98.09 98.09 98.09 e-np 12220 12245 12043 98.35 98.55 98.45 o 6349 6369 6199 97.33 97.64 97.48 e-vp 4768 4738 4665 98.46 97.84 98.15 i-vp 2602 2661 2535 95.26 97.43 96.33 e-adjp 384 371 332 89.49 86.46 87.95 i-pp 52 38 35 92.11 67.31 77.78 e-advp 822 816 724 88.73 88.08 88.40 i-advp 100 81 70 86.42 70.00 77.35 e-sbar 503 486 460 94.65 91.45 93.02 i-adjp 152 136 118 86.76 77.63 81.94 e-prt 126 129 121 93.80 96.03 94.90 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 15 15 100.00 62.50 76.92 e-conjp 16 8 8 100.00 50.00 66.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.31 72.48 75.28 Avg2. 46451 46451 45322 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12245 11790 96.28 96.48 96.38 pp 4633 4672 4573 97.88 98.70 98.29 vp 4768 4738 4558 96.20 95.60 95.90 sbar 503 486 455 93.62 90.46 92.01 adjp 384 371 318 85.71 82.81 84.24 advp 822 816 718 87.99 87.35 87.67 prt 126 129 121 93.80 96.03 94.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 8 8 100.00 50.00 66.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.15 77.74 80.36 Avg2. 23486 23475 22549 96.06 96.01 96.03 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 81 Log-likelihood = -71223.628745 Norm (log-likelihood gradient vector) = 2692.453992 Norm (lambda vector) = 217.326761 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4577 98.13 98.79 98.46 i-np 13660 13735 13413 97.66 98.19 97.92 e-np 12220 12206 12016 98.44 98.33 98.39 o 6349 6332 6173 97.49 97.23 97.36 e-vp 4768 4737 4663 98.44 97.80 98.12 i-vp 2602 2653 2532 95.44 97.31 96.37 e-adjp 384 373 334 89.54 86.98 88.24 i-pp 52 39 35 89.74 67.31 76.92 e-advp 822 813 723 88.93 87.96 88.44 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 496 468 94.35 93.04 93.69 i-adjp 152 140 120 85.71 78.95 82.19 e-prt 126 130 122 93.85 96.83 95.31 i-sbar 12 16 11 68.75 91.67 78.57 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.09 73.16 75.55 Avg2. 46451 46451 45290 97.50 97.50 97.50 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12206 11751 96.27 96.16 96.22 pp 4633 4664 4570 97.98 98.64 98.31 vp 4768 4737 4553 96.12 95.49 95.80 sbar 503 496 463 93.35 92.05 92.69 adjp 384 373 318 85.25 82.81 84.02 advp 822 813 717 88.19 87.23 87.71 prt 126 130 122 93.85 96.83 95.31 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.10 78.55 80.76 Avg2. 23486 23438 22511 96.04 95.85 95.95 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 82 Log-likelihood = -69446.216116 Norm (log-likelihood gradient vector) = 3298.463800 Norm (lambda vector) = 222.079725 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4578 98.05 98.81 98.43 i-np 13660 13672 13405 98.05 98.13 98.09 e-np 12220 12242 12044 98.38 98.56 98.47 o 6349 6359 6194 97.41 97.56 97.48 e-vp 4768 4738 4664 98.44 97.82 98.13 i-vp 2602 2654 2532 95.40 97.31 96.35 e-adjp 384 374 334 89.30 86.98 88.13 i-pp 52 38 35 92.11 67.31 77.78 e-advp 822 813 725 89.18 88.20 88.69 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 491 464 94.50 92.25 93.36 i-adjp 152 138 119 86.23 78.29 82.07 e-prt 126 130 122 93.85 96.83 95.31 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 16 16 100.00 66.67 80.00 e-conjp 16 9 9 100.00 56.25 72.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.49 73.18 75.74 Avg2. 46451 46451 45331 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12242 11792 96.32 96.50 96.41 pp 4633 4669 4571 97.90 98.66 98.28 vp 4768 4738 4555 96.14 95.53 95.83 sbar 503 491 459 93.48 91.25 92.35 adjp 384 374 319 85.29 83.07 84.17 advp 822 813 719 88.44 87.47 87.95 prt 126 130 122 93.85 96.83 95.31 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 9 9 100.00 56.25 72.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.14 78.56 80.78 Avg2. 23486 23476 22554 96.07 96.03 96.05 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 83 Log-likelihood = -68587.848165 Norm (log-likelihood gradient vector) = 1850.989905 Norm (lambda vector) = 223.363440 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4573 98.05 98.70 98.38 i-np 13660 13673 13409 98.07 98.16 98.12 e-np 12220 12237 12046 98.44 98.58 98.51 o 6349 6357 6195 97.45 97.57 97.51 e-vp 4768 4737 4662 98.42 97.78 98.10 i-vp 2602 2657 2532 95.30 97.31 96.29 e-adjp 384 374 334 89.30 86.98 88.13 i-pp 52 38 35 92.11 67.31 77.78 e-advp 822 814 726 89.19 88.32 88.75 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 494 464 93.93 92.25 93.08 i-adjp 152 138 120 86.96 78.95 82.76 e-prt 126 130 122 93.85 96.83 95.31 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 18 18 100.00 75.00 85.71 e-conjp 16 10 10 100.00 62.50 76.92 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.40 73.94 76.11 Avg2. 46451 46451 45336 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12237 11795 96.39 96.52 96.46 pp 4633 4664 4566 97.90 98.55 98.23 vp 4768 4737 4553 96.12 95.49 95.80 sbar 503 494 459 92.91 91.25 92.08 adjp 384 374 321 85.83 83.59 84.70 advp 822 814 720 88.45 87.59 88.02 prt 126 130 122 93.85 96.83 95.31 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 10 10 100.00 62.50 76.92 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.14 79.23 81.14 Avg2. 23486 23470 22554 96.10 96.03 96.06 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 84 Log-likelihood = -67887.889606 Norm (log-likelihood gradient vector) = 1633.714257 Norm (lambda vector) = 224.495075 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4574 98.03 98.73 98.38 i-np 13660 13684 13405 97.96 98.13 98.05 e-np 12220 12231 12039 98.43 98.52 98.47 o 6349 6355 6191 97.42 97.51 97.47 e-vp 4768 4734 4661 98.46 97.76 98.11 i-vp 2602 2657 2532 95.30 97.31 96.29 e-adjp 384 377 336 89.12 87.50 88.30 i-pp 52 38 35 92.11 67.31 77.78 e-advp 822 813 725 89.18 88.20 88.69 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 491 462 94.09 91.85 92.96 i-adjp 152 134 120 89.55 78.95 83.92 e-prt 126 130 122 93.85 96.83 95.31 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 20 20 100.00 83.33 90.91 e-conjp 16 11 11 100.00 68.75 81.48 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.52 74.67 76.54 Avg2. 46451 46451 45323 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11780 96.31 96.40 96.36 pp 4633 4666 4567 97.88 98.58 98.23 vp 4768 4734 4552 96.16 95.47 95.81 sbar 503 491 457 93.08 90.85 91.95 adjp 384 377 324 85.94 84.38 85.15 advp 822 813 719 88.44 87.47 87.95 prt 126 130 122 93.85 96.83 95.31 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 11 100.00 68.75 81.48 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 83.16 79.87 81.49 Avg2. 23486 23463 22540 96.07 95.97 96.02 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 85 Log-likelihood = -66740.591586 Norm (log-likelihood gradient vector) = 1892.339799 Norm (lambda vector) = 226.339261 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4651 4568 98.22 98.60 98.41 i-np 13660 13694 13406 97.90 98.14 98.02 e-np 12220 12224 12034 98.45 98.48 98.46 o 6349 6348 6186 97.45 97.43 97.44 e-vp 4768 4731 4659 98.48 97.71 98.09 i-vp 2602 2657 2534 95.37 97.39 96.37 e-adjp 384 379 336 88.65 87.50 88.07 i-pp 52 37 35 94.59 67.31 78.65 e-advp 822 806 720 89.33 87.59 88.45 i-advp 100 88 71 80.68 71.00 75.53 e-sbar 503 508 470 92.52 93.44 92.98 i-adjp 152 139 121 87.05 79.61 83.16 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 20 20 100.00 83.33 90.91 e-conjp 16 11 11 100.00 68.75 81.48 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 78.23 74.78 76.46 Avg2. 46451 46451 45313 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11772 96.30 96.33 96.32 pp 4633 4651 4561 98.06 98.45 98.26 vp 4768 4731 4554 96.26 95.51 95.88 sbar 503 508 465 91.54 92.45 91.99 adjp 384 379 324 85.49 84.38 84.93 advp 822 806 714 88.59 86.86 87.71 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 11 100.00 68.75 81.48 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.87 80.03 81.43 Avg2. 23486 23453 22532 96.07 95.94 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 86 Log-likelihood = -65526.056815 Norm (log-likelihood gradient vector) = 3852.663011 Norm (lambda vector) = 229.767398 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4574 98.07 98.73 98.40 i-np 13660 13753 13432 97.67 98.33 98.00 e-np 12220 12193 12019 98.57 98.36 98.46 o 6349 6318 6168 97.63 97.15 97.39 e-vp 4768 4731 4659 98.48 97.71 98.09 i-vp 2602 2650 2530 95.47 97.23 96.34 e-adjp 384 378 337 89.15 87.76 88.45 i-pp 52 37 35 94.59 67.31 78.65 e-advp 822 812 723 89.04 87.96 88.49 i-advp 100 88 70 79.55 70.00 74.47 e-sbar 503 495 465 93.94 92.45 93.19 i-adjp 152 141 124 87.94 81.58 84.64 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.64 74.80 76.19 Avg2. 46451 46451 45309 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11752 96.38 96.17 96.28 pp 4633 4664 4567 97.92 98.58 98.25 vp 4768 4731 4554 96.26 95.51 95.88 sbar 503 495 460 92.93 91.45 92.18 adjp 384 378 326 86.24 84.90 85.56 advp 822 812 717 88.30 87.23 87.76 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 80.02 81.10 Avg2. 23486 23428 22518 96.12 95.88 96.00 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 87 Log-likelihood = -63961.555078 Norm (log-likelihood gradient vector) = 2739.499232 Norm (lambda vector) = 231.378627 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 323 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4576 98.01 98.77 98.39 i-np 13660 13639 13392 98.19 98.04 98.11 e-np 12220 12246 12054 98.43 98.64 98.54 o 6349 6376 6204 97.30 97.72 97.51 e-vp 4768 4730 4659 98.50 97.71 98.10 i-vp 2602 2651 2530 95.44 97.23 96.33 e-adjp 384 378 336 88.89 87.50 88.19 i-pp 52 37 35 94.59 67.31 78.65 e-advp 822 806 719 89.21 87.47 88.33 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 492 463 94.11 92.05 93.07 i-adjp 152 150 125 83.33 82.24 82.78 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.51 74.80 76.13 Avg2. 46451 46451 45336 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12246 11806 96.41 96.61 96.51 pp 4633 4669 4569 97.86 98.62 98.24 vp 4768 4730 4553 96.26 95.49 95.87 sbar 503 492 458 93.09 91.05 92.06 adjp 384 378 324 85.71 84.38 85.04 advp 822 806 713 88.46 86.74 87.59 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.19 79.93 81.04 Avg2. 23486 23476 22565 96.12 96.08 96.10 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 88 Log-likelihood = -63231.945285 Norm (log-likelihood gradient vector) = 3311.653188 Norm (lambda vector) = 232.115562 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4576 98.07 98.77 98.42 i-np 13660 13700 13405 97.85 98.13 97.99 e-np 12220 12218 12029 98.45 98.44 98.45 o 6349 6347 6184 97.43 97.40 97.42 e-vp 4768 4730 4661 98.54 97.76 98.15 i-vp 2602 2651 2532 95.51 97.31 96.40 e-adjp 384 375 335 89.33 87.24 88.27 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 808 720 89.11 87.59 88.34 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 493 465 94.32 92.45 93.37 i-adjp 152 148 125 84.46 82.24 83.33 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.65 74.95 76.28 Avg2. 46451 46451 45312 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11770 96.33 96.32 96.33 pp 4633 4666 4569 97.92 98.62 98.27 vp 4768 4730 4556 96.32 95.55 95.94 sbar 503 493 460 93.31 91.45 92.37 adjp 384 375 322 85.87 83.85 84.85 advp 822 808 714 88.37 86.86 87.61 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.23 79.90 81.05 Avg2. 23486 23445 22533 96.11 95.94 96.03 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 89 Log-likelihood = -62647.213730 Norm (log-likelihood gradient vector) = 1793.175336 Norm (lambda vector) = 231.883766 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4577 98.18 98.79 98.48 i-np 13660 13732 13414 97.68 98.20 97.94 e-np 12220 12200 12019 98.52 98.36 98.44 o 6349 6333 6177 97.54 97.29 97.41 e-vp 4768 4731 4661 98.52 97.76 98.14 i-vp 2602 2653 2531 95.40 97.27 96.33 e-adjp 384 375 334 89.07 86.98 88.01 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 807 719 89.10 87.47 88.28 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 495 468 94.55 93.04 93.79 i-adjp 152 148 125 84.46 82.24 83.33 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.65 74.95 76.28 Avg2. 46451 46451 45305 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12200 11757 96.37 96.21 96.29 pp 4633 4662 4570 98.03 98.64 98.33 vp 4768 4731 4554 96.26 95.51 95.88 sbar 503 495 463 93.54 92.05 92.79 adjp 384 375 321 85.60 83.59 84.58 advp 822 807 713 88.35 86.74 87.54 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.23 79.91 81.05 Avg2. 23486 23425 22520 96.14 95.89 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 90 Log-likelihood = -62156.343678 Norm (log-likelihood gradient vector) = 1710.006084 Norm (lambda vector) = 231.983929 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4659 4576 98.22 98.77 98.49 i-np 13660 13751 13415 97.56 98.21 97.88 e-np 12220 12193 12013 98.52 98.31 98.41 o 6349 6324 6168 97.53 97.15 97.34 e-vp 4768 4738 4667 98.50 97.88 98.19 i-vp 2602 2645 2530 95.65 97.23 96.44 e-adjp 384 375 336 89.60 87.50 88.54 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 812 720 88.67 87.59 88.13 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 494 467 94.53 92.84 93.68 i-adjp 152 146 124 84.93 81.58 83.22 e-prt 126 133 123 92.48 97.62 94.98 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.59 74.52 76.02 Avg2. 46451 46451 45295 97.51 97.51 97.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11745 96.33 96.11 96.22 pp 4633 4659 4569 98.07 98.62 98.34 vp 4768 4738 4561 96.26 95.66 95.96 sbar 503 494 461 93.32 91.65 92.48 adjp 384 375 325 86.67 84.64 85.64 advp 822 812 714 87.93 86.86 87.39 prt 126 133 123 92.48 97.62 94.98 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.27 79.99 81.12 Avg2. 23486 23426 22517 96.12 95.87 96.00 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 91 Log-likelihood = -61301.980949 Norm (log-likelihood gradient vector) = 1805.052640 Norm (lambda vector) = 232.890634 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4647 4571 98.36 98.66 98.51 i-np 13660 13744 13414 97.60 98.20 97.90 e-np 12220 12201 12017 98.49 98.34 98.42 o 6349 6334 6177 97.52 97.29 97.41 e-vp 4768 4740 4668 98.48 97.90 98.19 i-vp 2602 2653 2533 95.48 97.35 96.40 e-adjp 384 370 334 90.27 86.98 88.59 i-pp 52 38 36 94.74 69.23 80.00 e-advp 822 805 716 88.94 87.10 88.01 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 496 468 94.35 93.04 93.69 i-adjp 152 146 124 84.93 81.58 83.22 e-prt 126 134 124 92.54 98.41 95.38 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.63 74.53 76.05 Avg2. 46451 46451 45302 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12201 11750 96.30 96.15 96.23 pp 4633 4647 4564 98.21 98.51 98.36 vp 4768 4740 4562 96.24 95.68 95.96 sbar 503 496 462 93.15 91.85 92.49 adjp 384 370 324 87.57 84.38 85.94 advp 822 805 710 88.20 86.37 87.28 prt 126 134 124 92.54 98.41 95.38 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.39 80.01 81.18 Avg2. 23486 23415 22515 96.16 95.87 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 92 Log-likelihood = -59685.090485 Norm (log-likelihood gradient vector) = 2334.200088 Norm (lambda vector) = 235.353468 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4688 4587 97.85 99.01 98.42 i-np 13660 13745 13398 97.48 98.08 97.78 e-np 12220 12206 12018 98.46 98.35 98.40 o 6349 6308 6160 97.65 97.02 97.34 e-vp 4768 4741 4667 98.44 97.88 98.16 i-vp 2602 2649 2532 95.58 97.31 96.44 e-adjp 384 375 337 89.87 87.76 88.80 i-pp 52 41 36 87.80 69.23 77.42 e-advp 822 813 717 88.19 87.23 87.71 i-advp 100 86 72 83.72 72.00 77.42 e-sbar 503 470 452 96.17 89.86 92.91 i-adjp 152 142 125 88.03 82.24 85.03 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 20 19 95.00 79.17 86.36 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.74 73.90 75.78 Avg2. 46451 46451 45273 97.46 97.46 97.46 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12206 11728 96.08 95.97 96.03 pp 4633 4688 4576 97.61 98.77 98.19 vp 4768 4741 4562 96.22 95.68 95.95 sbar 503 470 447 95.11 88.87 91.88 adjp 384 375 328 87.47 85.42 86.43 advp 822 813 712 87.58 86.62 87.09 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.57 79.93 81.23 Avg2. 23486 23448 22497 95.94 95.79 95.87 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 93 Log-likelihood = -60392.482373 Norm (log-likelihood gradient vector) = 7058.631506 Norm (lambda vector) = 239.404130 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4581 98.16 98.88 98.52 i-np 13660 13762 13414 97.47 98.20 97.83 e-np 12220 12197 12011 98.48 98.29 98.38 o 6349 6311 6162 97.64 97.05 97.35 e-vp 4768 4744 4669 98.42 97.92 98.17 i-vp 2602 2651 2532 95.51 97.31 96.40 e-adjp 384 369 335 90.79 87.24 88.98 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 808 717 88.74 87.23 87.98 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 485 461 95.05 91.65 93.32 i-adjp 152 145 124 85.52 81.58 83.50 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 19 18 94.74 75.00 83.72 e-conjp 16 11 10 90.91 62.50 74.07 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.71 73.74 75.67 Avg2. 46451 46451 45283 97.49 97.49 97.49 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12197 11738 96.24 96.06 96.15 pp 4633 4667 4572 97.96 98.68 98.32 vp 4768 4744 4563 96.18 95.70 95.94 sbar 503 485 456 94.02 90.66 92.31 adjp 384 369 325 88.08 84.64 86.32 advp 822 808 711 88.00 86.50 87.24 prt 126 133 124 93.23 98.41 95.75 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 10 90.91 62.50 74.07 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 79.31 80.86 Avg2. 23486 23424 22507 96.09 95.83 95.96 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 94 Log-likelihood = -58998.632799 Norm (log-likelihood gradient vector) = 2517.627913 Norm (lambda vector) = 237.057829 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4660 4579 98.26 98.83 98.55 i-np 13660 13721 13410 97.73 98.17 97.95 e-np 12220 12213 12030 98.50 98.45 98.47 o 6349 6334 6182 97.60 97.37 97.48 e-vp 4768 4744 4668 98.40 97.90 98.15 i-vp 2602 2653 2532 95.44 97.31 96.37 e-adjp 384 374 336 89.84 87.50 88.65 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 807 715 88.60 86.98 87.78 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 489 465 95.09 92.45 93.75 i-adjp 152 142 124 87.32 81.58 84.35 e-prt 126 134 125 93.28 99.21 96.15 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 19 18 94.74 75.00 83.72 e-conjp 16 11 10 90.91 62.50 74.07 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.49 73.84 75.62 Avg2. 46451 46451 45319 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11766 96.34 96.28 96.31 pp 4633 4660 4570 98.07 98.64 98.35 vp 4768 4744 4562 96.16 95.68 95.92 sbar 503 489 459 93.87 91.25 92.54 adjp 384 374 327 87.43 85.16 86.28 advp 822 807 709 87.86 86.25 87.05 prt 126 134 125 93.28 99.21 96.15 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 10 90.91 62.50 74.07 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.39 79.50 80.92 Avg2. 23486 23442 22536 96.14 95.96 96.05 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 95 Log-likelihood = -58161.273291 Norm (log-likelihood gradient vector) = 1388.571004 Norm (lambda vector) = 238.196154 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4656 4574 98.24 98.73 98.48 i-np 13660 13694 13402 97.87 98.11 97.99 e-np 12220 12224 12039 98.49 98.52 98.50 o 6349 6346 6190 97.54 97.50 97.52 e-vp 4768 4737 4661 98.40 97.76 98.07 i-vp 2602 2655 2530 95.29 97.23 96.25 e-adjp 384 377 337 89.39 87.76 88.57 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 811 717 88.41 87.23 87.81 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 496 467 94.15 92.84 93.49 i-adjp 152 141 123 87.23 80.92 83.96 e-prt 126 134 125 93.28 99.21 96.15 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 19 18 94.74 75.00 83.72 e-conjp 16 11 10 90.91 62.50 74.07 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.40 73.84 75.58 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11780 96.37 96.40 96.38 pp 4633 4656 4565 98.05 98.53 98.29 vp 4768 4737 4551 96.07 95.45 95.76 sbar 503 496 461 92.94 91.65 92.29 adjp 384 377 327 86.74 85.16 85.94 advp 822 811 711 87.67 86.50 87.08 prt 126 134 125 93.28 99.21 96.15 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 11 10 90.91 62.50 74.07 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.20 79.54 80.85 Avg2. 23486 23456 22538 96.09 95.96 96.02 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 96 Log-likelihood = -57456.703598 Norm (log-likelihood gradient vector) = 1284.751218 Norm (lambda vector) = 239.376938 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4578 98.16 98.81 98.48 i-np 13660 13687 13400 97.90 98.10 98.00 e-np 12220 12234 12044 98.45 98.56 98.50 o 6349 6345 6191 97.57 97.51 97.54 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2658 2538 95.49 97.54 96.50 e-adjp 384 377 337 89.39 87.76 88.57 i-pp 52 40 36 90.00 69.23 78.26 e-advp 822 809 721 89.12 87.71 88.41 i-advp 100 88 72 81.82 72.00 76.60 e-sbar 503 493 465 94.32 92.45 93.37 i-adjp 152 137 119 86.86 78.29 82.35 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 13 10 76.92 83.33 80.00 i-conjp 24 20 19 95.00 79.17 86.36 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.79 74.31 76.01 Avg2. 46451 46451 45335 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11789 96.36 96.47 96.42 pp 4633 4664 4569 97.96 98.62 98.29 vp 4768 4732 4558 96.32 95.60 95.96 sbar 503 493 460 93.31 91.45 92.37 adjp 384 377 326 86.47 84.90 85.68 advp 822 809 714 88.26 86.86 87.55 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.50 80.19 81.33 Avg2. 23486 23463 22560 96.15 96.06 96.10 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 97 Log-likelihood = -56570.928246 Norm (log-likelihood gradient vector) = 1463.313503 Norm (lambda vector) = 240.972772 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4649 4567 98.24 98.58 98.41 i-np 13660 13551 13329 98.36 97.58 97.97 e-np 12220 12291 12066 98.17 98.74 98.45 o 6349 6411 6215 96.94 97.89 97.41 e-vp 4768 4732 4659 98.46 97.71 98.08 i-vp 2602 2659 2536 95.37 97.46 96.41 e-adjp 384 375 338 90.13 88.02 89.06 i-pp 52 42 36 85.71 69.23 76.60 e-advp 822 814 719 88.33 87.47 87.90 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 506 470 92.89 93.44 93.16 i-adjp 152 146 122 83.56 80.26 81.88 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.07 74.70 75.87 Avg2. 46451 46451 45303 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12291 11810 96.09 96.64 96.36 pp 4633 4649 4557 98.02 98.36 98.19 vp 4768 4732 4555 96.26 95.53 95.89 sbar 503 506 463 91.50 92.05 91.77 adjp 384 375 324 86.40 84.38 85.38 advp 822 814 712 87.47 86.62 87.04 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.27 80.78 81.52 Avg2. 23486 23522 22566 95.94 96.08 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 98 Log-likelihood = -56057.961550 Norm (log-likelihood gradient vector) = 6390.883888 Norm (lambda vector) = 243.798992 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4575 98.15 98.75 98.45 i-np 13660 13706 13406 97.81 98.14 97.98 e-np 12220 12219 12034 98.49 98.48 98.48 o 6349 6330 6177 97.58 97.29 97.44 e-vp 4768 4730 4660 98.52 97.73 98.13 i-vp 2602 2658 2536 95.41 97.46 96.43 e-adjp 384 374 337 90.11 87.76 88.92 i-pp 52 43 36 83.72 69.23 75.79 e-advp 822 813 720 88.56 87.59 88.07 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 497 466 93.76 92.64 93.20 i-adjp 152 146 122 83.56 80.26 81.88 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.33 74.65 75.97 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11778 96.39 96.38 96.39 pp 4633 4661 4564 97.92 98.51 98.21 vp 4768 4730 4557 96.34 95.57 95.96 sbar 503 497 460 92.56 91.45 92.00 adjp 384 374 323 86.36 84.11 85.22 advp 822 813 713 87.70 86.74 87.22 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.43 80.70 81.55 Avg2. 23486 23449 22540 96.12 95.97 96.05 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 99 Log-likelihood = -54865.238724 Norm (log-likelihood gradient vector) = 2017.869984 Norm (lambda vector) = 245.181757 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4577 98.09 98.79 98.44 i-np 13660 13708 13402 97.77 98.11 97.94 e-np 12220 12218 12031 98.47 98.45 98.46 o 6349 6326 6171 97.55 97.20 97.37 e-vp 4768 4731 4660 98.50 97.73 98.12 i-vp 2602 2658 2536 95.41 97.46 96.43 e-adjp 384 374 337 90.11 87.76 88.92 i-pp 52 44 36 81.82 69.23 75.00 e-advp 822 813 720 88.56 87.59 88.07 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 493 464 94.12 92.25 93.17 i-adjp 152 146 122 83.56 80.26 81.88 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.25 74.63 75.91 Avg2. 46451 46451 45302 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11770 96.33 96.32 96.33 pp 4633 4666 4566 97.86 98.55 98.20 vp 4768 4731 4557 96.32 95.57 95.95 sbar 503 493 458 92.90 91.05 91.97 adjp 384 374 323 86.36 84.11 85.22 advp 822 813 713 87.70 86.74 87.22 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.45 80.66 81.54 Avg2. 23486 23450 22532 96.09 95.94 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 100 Log-likelihood = -54685.302979 Norm (log-likelihood gradient vector) = 1544.288882 Norm (lambda vector) = 244.616815 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4577 98.11 98.79 98.45 i-np 13660 13678 13398 97.95 98.08 98.02 e-np 12220 12237 12042 98.41 98.54 98.47 o 6349 6349 6185 97.42 97.42 97.42 e-vp 4768 4728 4662 98.60 97.78 98.19 i-vp 2602 2655 2536 95.52 97.46 96.48 e-adjp 384 372 336 90.32 87.50 88.89 i-pp 52 43 36 83.72 69.23 75.79 e-advp 822 814 722 88.70 87.83 88.26 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 491 464 94.50 92.25 93.36 i-adjp 152 146 122 83.56 80.26 81.88 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 11 9 81.82 75.00 78.26 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.73 74.64 76.16 Avg2. 46451 46451 45326 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12237 11785 96.31 96.44 96.37 pp 4633 4665 4567 97.90 98.58 98.24 vp 4768 4728 4559 96.43 95.62 96.02 sbar 503 491 459 93.48 91.25 92.35 adjp 384 372 322 86.56 83.85 85.19 advp 822 814 715 87.84 86.98 87.41 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.55 80.69 81.61 Avg2. 23486 23462 22552 96.12 96.02 96.07 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 101 Log-likelihood = -53938.244575 Norm (log-likelihood gradient vector) = 1137.276268 Norm (lambda vector) = 245.466908 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4579 98.18 98.83 98.50 i-np 13660 13692 13401 97.87 98.10 97.99 e-np 12220 12235 12042 98.42 98.54 98.48 o 6349 6349 6187 97.45 97.45 97.45 e-vp 4768 4733 4665 98.56 97.84 98.20 i-vp 2602 2642 2526 95.61 97.08 96.34 e-adjp 384 373 336 90.08 87.50 88.77 i-pp 52 42 36 85.71 69.23 76.60 e-advp 822 812 720 88.67 87.59 88.13 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 147 122 82.99 80.26 81.61 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.52 74.59 76.03 Avg2. 46451 46451 45324 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12235 11784 96.31 96.43 96.37 pp 4633 4664 4569 97.96 98.62 98.29 vp 4768 4733 4559 96.32 95.62 95.97 sbar 503 490 459 93.67 91.25 92.45 adjp 384 373 322 86.33 83.85 85.07 advp 822 812 713 87.81 86.74 87.27 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.54 80.59 81.55 Avg2. 23486 23461 22550 96.12 96.01 96.07 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 102 Log-likelihood = -53242.529023 Norm (log-likelihood gradient vector) = 1461.533488 Norm (lambda vector) = 246.622597 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4579 98.14 98.83 98.48 i-np 13660 13674 13383 97.87 97.97 97.92 e-np 12220 12249 12041 98.30 98.54 98.42 o 6349 6355 6181 97.26 97.35 97.31 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2644 2528 95.61 97.16 96.38 e-adjp 384 376 337 89.63 87.76 88.68 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 806 717 88.96 87.23 88.08 i-advp 100 82 69 84.15 69.00 75.82 e-sbar 503 488 463 94.88 92.05 93.44 i-adjp 152 148 121 81.76 79.61 80.67 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.39 74.33 75.83 Avg2. 46451 46451 45292 97.50 97.50 97.50 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12249 11772 96.11 96.33 96.22 pp 4633 4666 4569 97.92 98.62 98.27 vp 4768 4734 4560 96.32 95.64 95.98 sbar 503 488 457 93.65 90.85 92.23 adjp 384 376 322 85.64 83.85 84.74 advp 822 806 709 87.97 86.25 87.10 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 80.50 81.46 Avg2. 23486 23473 22533 96.00 95.94 95.97 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 103 Log-likelihood = -52336.440338 Norm (log-likelihood gradient vector) = 3029.869774 Norm (lambda vector) = 249.694919 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4578 98.16 98.81 98.48 i-np 13660 13672 13396 97.98 98.07 98.02 e-np 12220 12248 12048 98.37 98.59 98.48 o 6349 6354 6188 97.39 97.46 97.43 e-vp 4768 4733 4663 98.52 97.80 98.16 i-vp 2602 2650 2531 95.51 97.27 96.38 e-adjp 384 376 337 89.63 87.76 88.68 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 806 716 88.83 87.10 87.96 i-advp 100 81 69 85.19 69.00 76.24 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 148 121 81.76 79.61 80.67 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.45 74.36 75.87 Avg2. 46451 46451 45321 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12248 11790 96.26 96.48 96.37 pp 4633 4664 4568 97.94 98.60 98.27 vp 4768 4733 4558 96.30 95.60 95.95 sbar 503 490 459 93.67 91.25 92.45 adjp 384 376 322 85.64 83.85 84.74 advp 822 806 708 87.84 86.13 86.98 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 80.53 81.49 Avg2. 23486 23471 22549 96.07 96.01 96.04 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 104 Log-likelihood = -51555.780625 Norm (log-likelihood gradient vector) = 1415.983706 Norm (lambda vector) = 250.678125 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4578 98.20 98.81 98.50 i-np 13660 13702 13406 97.84 98.14 97.99 e-np 12220 12231 12041 98.45 98.54 98.49 o 6349 6339 6184 97.55 97.40 97.48 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2649 2531 95.55 97.27 96.40 e-adjp 384 380 337 88.68 87.76 88.22 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 807 716 88.72 87.10 87.91 i-advp 100 81 69 85.19 69.00 76.24 e-sbar 503 492 466 94.72 92.64 93.67 i-adjp 152 144 121 84.03 79.61 81.76 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.66 74.89 76.25 Avg2. 46451 46451 45324 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11786 96.36 96.45 96.41 pp 4633 4662 4568 97.98 98.60 98.29 vp 4768 4734 4560 96.32 95.64 95.98 sbar 503 492 460 93.50 91.45 92.46 adjp 384 380 323 85.00 84.11 84.55 advp 822 807 708 87.73 86.13 86.92 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.44 81.20 81.82 Avg2. 23486 23461 22550 96.12 96.01 96.07 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 105 Log-likelihood = -50935.660612 Norm (log-likelihood gradient vector) = 1152.812657 Norm (lambda vector) = 252.109007 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4579 98.20 98.83 98.52 i-np 13660 13727 13418 97.75 98.23 97.99 e-np 12220 12217 12031 98.48 98.45 98.47 o 6349 6330 6179 97.61 97.32 97.47 e-vp 4768 4729 4664 98.63 97.82 98.22 i-vp 2602 2648 2532 95.62 97.31 96.46 e-adjp 384 381 339 88.98 88.28 88.63 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 809 717 88.63 87.23 87.92 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 491 465 94.70 92.45 93.56 i-adjp 152 143 121 84.62 79.61 82.03 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 12 9 75.00 75.00 75.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.60 74.91 76.23 Avg2. 46451 46451 45325 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11778 96.41 96.38 96.39 pp 4633 4663 4569 97.98 98.62 98.30 vp 4768 4729 4561 96.45 95.66 96.05 sbar 503 491 459 93.48 91.25 92.35 adjp 384 381 325 85.30 84.64 84.97 advp 822 809 709 87.64 86.25 86.94 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.48 81.25 81.86 Avg2. 23486 23445 22546 96.17 96.00 96.08 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 106 Log-likelihood = -50417.583971 Norm (log-likelihood gradient vector) = 1383.620693 Norm (lambda vector) = 253.415646 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4582 98.16 98.90 98.53 i-np 13660 13650 13392 98.11 98.04 98.07 e-np 12220 12251 12050 98.36 98.61 98.48 o 6349 6367 6200 97.38 97.65 97.51 e-vp 4768 4729 4662 98.58 97.78 98.18 i-vp 2602 2650 2532 95.55 97.31 96.42 e-adjp 384 379 337 88.92 87.76 88.34 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 83 69 83.13 69.00 75.41 e-sbar 503 489 465 95.09 92.45 93.75 i-adjp 152 142 120 84.51 78.95 81.63 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.31 74.88 76.07 Avg2. 46451 46451 45339 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12251 11802 96.33 96.58 96.46 pp 4633 4668 4572 97.94 98.68 98.31 vp 4768 4729 4559 96.41 95.62 96.01 sbar 503 489 458 93.66 91.05 92.34 adjp 384 379 322 84.96 83.85 84.40 advp 822 812 711 87.56 86.50 87.03 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.44 81.19 81.81 Avg2. 23486 23483 22569 96.11 96.10 96.10 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 107 Log-likelihood = -49595.207704 Norm (log-likelihood gradient vector) = 2863.341603 Norm (lambda vector) = 256.156558 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4583 98.18 98.92 98.55 i-np 13660 13798 13435 97.37 98.35 97.86 e-np 12220 12187 12007 98.52 98.26 98.39 o 6349 6301 6156 97.70 96.96 97.33 e-vp 4768 4731 4662 98.54 97.78 98.16 i-vp 2602 2651 2535 95.62 97.43 96.52 e-adjp 384 369 333 90.24 86.72 88.45 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 812 723 89.04 87.96 88.49 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 488 464 95.08 92.25 93.64 i-adjp 152 133 115 86.47 75.66 80.70 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.60 74.70 76.12 Avg2. 46451 46451 45294 97.51 97.51 97.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12187 11742 96.35 96.09 96.22 pp 4633 4668 4573 97.96 98.70 98.33 vp 4768 4731 4562 96.43 95.68 96.05 sbar 503 488 457 93.65 90.85 92.23 adjp 384 369 318 86.18 82.81 84.46 advp 822 812 715 88.05 86.98 87.52 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.61 81.08 81.84 Avg2. 23486 23410 22512 96.16 95.85 96.01 Current max chunk-based F1: 96.13 (iteration 76) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 108 Log-likelihood = -48956.416054 Norm (log-likelihood gradient vector) = 4157.154279 Norm (lambda vector) = 259.000897 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4583 98.22 98.92 98.57 i-np 13660 13717 13420 97.83 98.24 98.04 e-np 12220 12219 12034 98.49 98.48 98.48 o 6349 6339 6185 97.57 97.42 97.49 e-vp 4768 4733 4665 98.56 97.84 98.20 i-vp 2602 2644 2530 95.69 97.23 96.45 e-adjp 384 378 337 89.15 87.76 88.45 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 813 722 88.81 87.83 88.32 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 489 465 95.09 92.45 93.75 i-adjp 152 138 118 85.51 77.63 81.38 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.41 74.87 76.12 Avg2. 46451 46451 45340 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11785 96.45 96.44 96.44 pp 4633 4666 4573 98.01 98.70 98.35 vp 4768 4733 4565 96.45 95.74 96.10 sbar 503 489 458 93.66 91.05 92.34 adjp 384 378 321 84.92 83.59 84.25 advp 822 813 714 87.82 86.86 87.34 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.48 81.21 81.84 Avg2. 23486 23453 22561 96.20 96.06 96.13 Current max chunk-based F1: 96.13 (iteration 108) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 109 Log-likelihood = -48632.539119 Norm (log-likelihood gradient vector) = 1463.395466 Norm (lambda vector) = 258.128975 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4582 98.20 98.90 98.55 i-np 13660 13683 13408 97.99 98.16 98.07 e-np 12220 12235 12042 98.42 98.54 98.48 o 6349 6356 6195 97.47 97.57 97.52 e-vp 4768 4735 4669 98.61 97.92 98.26 i-vp 2602 2640 2530 95.83 97.23 96.53 e-adjp 384 377 337 89.39 87.76 88.57 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 813 722 88.81 87.83 88.32 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 141 121 85.82 79.61 82.59 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.43 74.98 76.19 Avg2. 46451 46451 45352 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12235 11795 96.40 96.52 96.46 pp 4633 4666 4572 97.99 98.68 98.33 vp 4768 4735 4570 96.52 95.85 96.18 sbar 503 490 458 93.47 91.05 92.25 adjp 384 377 323 85.68 84.11 84.89 advp 822 813 714 87.82 86.86 87.34 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.54 81.27 81.90 Avg2. 23486 23471 22577 96.19 96.13 96.16 Current max chunk-based F1: 96.16 (iteration 109) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 110 Log-likelihood = -48461.556701 Norm (log-likelihood gradient vector) = 1227.039803 Norm (lambda vector) = 257.930311 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4582 98.20 98.90 98.55 i-np 13660 13672 13408 98.07 98.16 98.11 e-np 12220 12240 12047 98.42 98.58 98.50 o 6349 6361 6200 97.47 97.65 97.56 e-vp 4768 4735 4669 98.61 97.92 98.26 i-vp 2602 2639 2530 95.87 97.23 96.55 e-adjp 384 379 337 88.92 87.76 88.34 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 815 723 88.71 87.96 88.33 i-advp 100 83 70 84.34 70.00 76.50 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 140 119 85.00 78.29 81.51 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.42 74.93 76.15 Avg2. 46451 46451 45361 97.65 97.65 97.65 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11804 96.44 96.60 96.52 pp 4633 4666 4572 97.99 98.68 98.33 vp 4768 4735 4571 96.54 95.87 96.20 sbar 503 490 458 93.47 91.05 92.25 adjp 384 379 322 84.96 83.85 84.40 advp 822 815 715 87.73 86.98 87.35 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 81.27 81.86 Avg2. 23486 23480 22587 96.20 96.17 96.18 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 111 Log-likelihood = -48189.580420 Norm (log-likelihood gradient vector) = 1472.942201 Norm (lambda vector) = 258.392167 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4580 98.20 98.86 98.53 i-np 13660 13665 13398 98.05 98.08 98.06 e-np 12220 12246 12048 98.38 98.59 98.49 o 6349 6366 6200 97.39 97.65 97.52 e-vp 4768 4735 4668 98.59 97.90 98.24 i-vp 2602 2642 2530 95.76 97.23 96.49 e-adjp 384 377 335 88.86 87.24 88.04 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 812 721 88.79 87.71 88.25 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 492 465 94.51 92.45 93.47 i-adjp 152 139 118 84.89 77.63 81.10 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.43 74.85 76.12 Avg2. 46451 46451 45344 97.62 97.62 97.62 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12246 11800 96.36 96.56 96.46 pp 4633 4664 4570 97.98 98.64 98.31 vp 4768 4735 4568 96.47 95.81 96.14 sbar 503 492 458 93.09 91.05 92.06 adjp 384 377 320 84.88 83.33 84.10 advp 822 812 713 87.81 86.74 87.27 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.41 81.18 81.79 Avg2. 23486 23481 22574 96.14 96.12 96.13 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 112 Log-likelihood = -47549.896042 Norm (log-likelihood gradient vector) = 1699.788175 Norm (lambda vector) = 259.893145 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4645 4570 98.39 98.64 98.51 i-np 13660 13665 13386 97.96 97.99 97.98 e-np 12220 12243 12042 98.36 98.54 98.45 o 6349 6371 6197 97.27 97.61 97.44 e-vp 4768 4736 4668 98.56 97.90 98.23 i-vp 2602 2646 2531 95.65 97.27 96.46 e-adjp 384 374 335 89.57 87.24 88.39 i-pp 52 39 35 89.74 67.31 76.92 e-advp 822 813 722 88.81 87.83 88.32 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 509 470 92.34 93.44 92.89 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.59 75.24 76.40 Avg2. 46451 46451 45319 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12243 11784 96.25 96.43 96.34 pp 4633 4645 4560 98.17 98.42 98.30 vp 4768 4736 4569 96.47 95.83 96.15 sbar 503 509 464 91.16 92.25 91.70 adjp 384 374 320 85.56 83.33 84.43 advp 822 813 714 87.82 86.86 87.34 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 81.28 81.75 Avg2. 23486 23476 22556 96.08 96.04 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 113 Log-likelihood = -46543.133542 Norm (log-likelihood gradient vector) = 3740.237535 Norm (lambda vector) = 265.167914 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4579 98.24 98.83 98.54 i-np 13660 13678 13399 97.96 98.09 98.02 e-np 12220 12236 12043 98.42 98.55 98.49 o 6349 6361 6195 97.39 97.57 97.48 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2649 2528 95.43 97.16 96.29 e-adjp 384 374 334 89.30 86.98 88.13 i-pp 52 39 35 89.74 67.31 76.92 e-advp 822 817 723 88.49 87.96 88.22 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 494 465 94.13 92.45 93.28 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.71 75.18 76.43 Avg2. 46451 46451 45328 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12236 11788 96.34 96.46 96.40 pp 4633 4661 4569 98.03 98.62 98.32 vp 4768 4734 4557 96.26 95.57 95.92 sbar 503 494 459 92.91 91.25 92.08 adjp 384 374 319 85.29 83.07 84.17 advp 822 817 715 87.52 86.98 87.25 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.39 81.16 81.77 Avg2. 23486 23471 22552 96.08 96.02 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 114 Log-likelihood = -45741.335117 Norm (log-likelihood gradient vector) = 1541.465471 Norm (lambda vector) = 265.909853 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4582 98.24 98.90 98.57 i-np 13660 13712 13415 97.83 98.21 98.02 e-np 12220 12219 12037 98.51 98.50 98.51 o 6349 6344 6189 97.56 97.48 97.52 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2650 2529 95.43 97.19 96.31 e-adjp 384 373 334 89.54 86.98 88.24 i-pp 52 39 35 89.74 67.31 76.92 e-advp 822 815 721 88.47 87.71 88.09 i-advp 100 83 70 84.34 70.00 76.50 e-sbar 503 492 465 94.51 92.45 93.47 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.69 75.18 76.41 Avg2. 46451 46451 45334 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11784 96.44 96.43 96.44 pp 4633 4664 4572 98.03 98.68 98.35 vp 4768 4734 4558 96.28 95.60 95.94 sbar 503 492 459 93.29 91.25 92.26 adjp 384 373 319 85.52 83.07 84.28 advp 822 815 713 87.48 86.74 87.11 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 81.14 81.79 Avg2. 23486 23452 22550 96.15 96.01 96.08 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 115 Log-likelihood = -45376.935915 Norm (log-likelihood gradient vector) = 1033.412578 Norm (lambda vector) = 266.565132 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4582 98.22 98.90 98.56 i-np 13660 13741 13422 97.68 98.26 97.97 e-np 12220 12206 12026 98.53 98.41 98.47 o 6349 6334 6178 97.54 97.31 97.42 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2643 2525 95.54 97.04 96.28 e-adjp 384 372 333 89.52 86.72 88.10 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 815 720 88.34 87.59 87.97 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 489 464 94.89 92.25 93.55 i-adjp 152 133 115 86.47 75.66 80.70 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.58 75.55 76.55 Avg2. 46451 46451 45313 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12206 11770 96.43 96.32 96.37 pp 4633 4665 4572 98.01 98.68 98.34 vp 4768 4734 4558 96.28 95.60 95.94 sbar 503 489 459 93.87 91.25 92.54 adjp 384 372 318 85.48 82.81 84.13 advp 822 815 712 87.36 86.62 86.99 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.50 81.17 81.83 Avg2. 23486 23437 22535 96.15 95.95 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 116 Log-likelihood = -44853.114450 Norm (log-likelihood gradient vector) = 1184.554050 Norm (lambda vector) = 268.130721 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4581 98.22 98.88 98.55 i-np 13660 13744 13416 97.61 98.21 97.91 e-np 12220 12204 12023 98.52 98.39 98.45 o 6349 6337 6176 97.46 97.28 97.37 e-vp 4768 4736 4666 98.52 97.86 98.19 i-vp 2602 2640 2524 95.61 97.00 96.30 e-adjp 384 373 334 89.54 86.98 88.24 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 813 717 88.19 87.23 87.71 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 489 464 94.89 92.25 93.55 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.57 75.51 76.52 Avg2. 46451 46451 45299 97.52 97.52 97.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12204 11762 96.38 96.25 96.32 pp 4633 4664 4571 98.01 98.66 98.33 vp 4768 4736 4560 96.28 95.64 95.96 sbar 503 489 459 93.87 91.25 92.54 adjp 384 373 318 85.25 82.81 84.02 advp 822 813 709 87.21 86.25 86.73 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.45 81.13 81.79 Avg2. 23486 23435 22525 96.12 95.91 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 117 Log-likelihood = -44279.227417 Norm (log-likelihood gradient vector) = 1591.191964 Norm (lambda vector) = 269.978221 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4674 4587 98.14 99.01 98.57 i-np 13660 13676 13396 97.95 98.07 98.01 e-np 12220 12241 12045 98.40 98.57 98.48 o 6349 6354 6187 97.37 97.45 97.41 e-vp 4768 4739 4670 98.54 97.94 98.24 i-vp 2602 2643 2532 95.80 97.31 96.55 e-adjp 384 371 335 90.30 87.24 88.74 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 815 722 88.59 87.83 88.21 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 483 461 95.45 91.65 93.51 i-adjp 152 136 114 83.82 75.00 79.17 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.54 75.56 76.54 Avg2. 46451 46451 45333 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12241 11784 96.27 96.43 96.35 pp 4633 4674 4577 97.92 98.79 98.36 vp 4768 4739 4573 96.50 95.91 96.20 sbar 503 483 456 94.41 90.66 92.49 adjp 384 371 318 85.71 82.81 84.24 advp 822 815 714 87.61 86.86 87.23 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.60 81.19 81.89 Avg2. 23486 23479 22568 96.12 96.09 96.11 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 118 Log-likelihood = -43607.826910 Norm (log-likelihood gradient vector) = 2434.443098 Norm (lambda vector) = 273.929104 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4580 98.24 98.86 98.55 i-np 13660 13819 13442 97.27 98.40 97.83 e-np 12220 12175 12002 98.58 98.22 98.40 o 6349 6297 6151 97.68 96.88 97.28 e-vp 4768 4738 4669 98.54 97.92 98.23 i-vp 2602 2637 2527 95.83 97.12 96.47 e-adjp 384 367 331 90.19 86.20 88.15 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 817 721 88.25 87.71 87.98 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 488 464 95.08 92.25 93.64 i-adjp 152 132 111 84.09 73.03 78.17 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.31 74.97 76.12 Avg2. 46451 46451 45281 97.48 97.48 97.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12175 11727 96.32 95.97 96.14 pp 4633 4662 4570 98.03 98.64 98.33 vp 4768 4738 4571 96.48 95.87 96.17 sbar 503 488 458 93.85 91.05 92.43 adjp 384 367 314 85.56 81.77 83.62 advp 822 817 714 87.39 86.86 87.13 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.52 81.06 81.78 Avg2. 23486 23403 22500 96.14 95.80 95.97 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 119 Log-likelihood = -43162.740548 Norm (log-likelihood gradient vector) = 4023.590263 Norm (lambda vector) = 274.892914 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4581 98.28 98.88 98.58 i-np 13660 13748 13424 97.64 98.27 97.96 e-np 12220 12207 12026 98.52 98.41 98.46 o 6349 6329 6174 97.55 97.24 97.40 e-vp 4768 4739 4670 98.54 97.94 98.24 i-vp 2602 2637 2527 95.83 97.12 96.47 e-adjp 384 369 333 90.24 86.72 88.45 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 818 722 88.26 87.83 88.05 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 134 112 83.58 73.68 78.32 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.34 75.07 76.19 Avg2. 46451 46451 45317 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12207 11766 96.39 96.28 96.34 pp 4633 4661 4571 98.07 98.66 98.36 vp 4768 4739 4572 96.48 95.89 96.18 sbar 503 490 459 93.67 91.25 92.45 adjp 384 369 316 85.64 82.29 83.93 advp 822 818 714 87.29 86.86 87.07 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.51 81.17 81.83 Avg2. 23486 23440 22544 96.18 95.99 96.08 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 120 Log-likelihood = -42834.098748 Norm (log-likelihood gradient vector) = 1503.586837 Norm (lambda vector) = 274.399104 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4656 4576 98.28 98.77 98.53 i-np 13660 13700 13412 97.90 98.18 98.04 e-np 12220 12231 12040 98.44 98.53 98.48 o 6349 6354 6188 97.39 97.46 97.43 e-vp 4768 4741 4671 98.52 97.97 98.24 i-vp 2602 2637 2527 95.83 97.12 96.47 e-adjp 384 369 333 90.24 86.72 88.45 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 819 723 88.28 87.96 88.12 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 491 465 94.70 92.45 93.56 i-adjp 152 134 112 83.58 73.68 78.32 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 22 95.65 91.67 93.62 e-conjp 16 14 13 92.86 81.25 86.67 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.33 75.08 76.19 Avg2. 46451 46451 45330 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11787 96.37 96.46 96.41 pp 4633 4656 4566 98.07 98.55 98.31 vp 4768 4741 4572 96.44 95.89 96.16 sbar 503 491 459 93.48 91.25 92.35 adjp 384 369 316 85.64 82.29 83.93 advp 822 819 715 87.30 86.98 87.14 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 13 92.86 81.25 86.67 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.48 81.19 81.83 Avg2. 23486 23463 22561 96.16 96.06 96.11 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 121 Log-likelihood = -42639.081601 Norm (log-likelihood gradient vector) = 915.144404 Norm (lambda vector) = 274.697380 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4655 4575 98.28 98.75 98.51 i-np 13660 13689 13411 97.97 98.18 98.07 e-np 12220 12236 12044 98.43 98.56 98.50 o 6349 6362 6196 97.39 97.59 97.49 e-vp 4768 4742 4671 98.50 97.97 98.23 i-vp 2602 2639 2527 95.76 97.12 96.43 e-adjp 384 369 331 89.70 86.20 87.92 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 819 724 88.40 88.08 88.24 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 493 465 94.32 92.45 93.37 i-adjp 152 130 112 86.15 73.68 79.43 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.38 74.55 75.94 Avg2. 46451 46451 45337 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12236 11796 96.40 96.53 96.47 pp 4633 4655 4565 98.07 98.53 98.30 vp 4768 4742 4571 96.39 95.87 96.13 sbar 503 493 459 93.10 91.25 92.17 adjp 384 369 315 85.37 82.03 83.67 advp 822 819 716 87.42 87.10 87.26 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.38 80.55 81.45 Avg2. 23486 23469 22567 96.16 96.09 96.12 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 122 Log-likelihood = -42439.665248 Norm (log-likelihood gradient vector) = 1198.329474 Norm (lambda vector) = 275.316341 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4651 4573 98.32 98.70 98.51 i-np 13660 13681 13407 98.00 98.15 98.07 e-np 12220 12238 12045 98.42 98.57 98.50 o 6349 6368 6199 97.35 97.64 97.49 e-vp 4768 4747 4672 98.42 97.99 98.20 i-vp 2602 2635 2525 95.83 97.04 96.43 e-adjp 384 371 332 89.49 86.46 87.95 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 816 723 88.60 87.96 88.28 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 494 466 94.33 92.64 93.48 i-adjp 152 133 114 85.71 75.00 80.00 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.38 74.62 75.97 Avg2. 46451 46451 45337 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12238 11798 96.40 96.55 96.48 pp 4633 4651 4563 98.11 98.49 98.30 vp 4768 4747 4572 96.31 95.89 96.10 sbar 503 494 460 93.12 91.45 92.28 adjp 384 371 317 85.44 82.55 83.97 advp 822 816 715 87.62 86.98 87.30 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.33 80.61 81.46 Avg2. 23486 23473 22570 96.15 96.10 96.13 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 123 Log-likelihood = -41934.843364 Norm (log-likelihood gradient vector) = 1440.390684 Norm (lambda vector) = 277.021119 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4651 4575 98.37 98.75 98.56 i-np 13660 13671 13396 97.99 98.07 98.03 e-np 12220 12242 12045 98.39 98.57 98.48 o 6349 6372 6198 97.27 97.62 97.45 e-vp 4768 4742 4669 98.46 97.92 98.19 i-vp 2602 2638 2523 95.64 96.96 96.30 e-adjp 384 370 332 89.73 86.46 88.06 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 820 722 88.05 87.83 87.94 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 496 468 94.35 93.04 93.69 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.37 74.63 75.97 Avg2. 46451 46451 45323 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12242 11793 96.33 96.51 96.42 pp 4633 4651 4565 98.15 98.53 98.34 vp 4768 4742 4561 96.18 95.66 95.92 sbar 503 496 462 93.15 91.85 92.49 adjp 384 370 317 85.68 82.55 84.08 advp 822 820 714 87.07 86.86 86.97 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.36 80.62 81.48 Avg2. 23486 23476 22557 96.09 96.04 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 124 Log-likelihood = -41091.153788 Norm (log-likelihood gradient vector) = 1867.761760 Norm (lambda vector) = 279.570644 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4674 4583 98.05 98.92 98.49 i-np 13660 13732 13413 97.68 98.19 97.93 e-np 12220 12213 12026 98.47 98.41 98.44 o 6349 6332 6176 97.54 97.28 97.41 e-vp 4768 4738 4668 98.52 97.90 98.21 i-vp 2602 2645 2527 95.54 97.12 96.32 e-adjp 384 373 333 89.28 86.72 87.98 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 813 721 88.68 87.71 88.20 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 481 458 95.22 91.05 93.09 i-adjp 152 133 114 85.71 75.00 80.00 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.37 74.53 75.93 Avg2. 46451 46451 45300 97.52 97.52 97.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11766 96.34 96.28 96.31 pp 4633 4674 4573 97.84 98.70 98.27 vp 4768 4738 4561 96.26 95.66 95.96 sbar 503 481 452 93.97 89.86 91.87 adjp 384 373 317 84.99 82.55 83.75 advp 822 813 713 87.70 86.74 87.22 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.41 80.40 81.39 Avg2. 23486 23447 22527 96.08 95.92 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 125 Log-likelihood = -40945.302953 Norm (log-likelihood gradient vector) = 4115.568177 Norm (lambda vector) = 284.895117 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4579 98.22 98.83 98.53 i-np 13660 13684 13403 97.95 98.12 98.03 e-np 12220 12234 12040 98.41 98.53 98.47 o 6349 6360 6193 97.37 97.54 97.46 e-vp 4768 4739 4666 98.46 97.86 98.16 i-vp 2602 2642 2525 95.57 97.04 96.30 e-adjp 384 372 332 89.25 86.46 87.83 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 822 724 88.08 88.08 88.08 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 488 463 94.88 92.05 93.44 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.41 74.59 75.98 Avg2. 46451 46451 45320 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11787 96.35 96.46 96.40 pp 4633 4662 4569 98.01 98.62 98.31 vp 4768 4739 4559 96.20 95.62 95.91 sbar 503 488 457 93.65 90.85 92.23 adjp 384 372 317 85.22 82.55 83.86 advp 822 822 716 87.10 87.10 87.10 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.35 80.54 81.44 Avg2. 23486 23472 22550 96.07 96.01 96.04 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 126 Log-likelihood = -40526.468921 Norm (log-likelihood gradient vector) = 2048.710383 Norm (lambda vector) = 282.000667 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4582 98.31 98.90 98.60 i-np 13660 13700 13407 97.86 98.15 98.00 e-np 12220 12224 12034 98.45 98.48 98.46 o 6349 6356 6188 97.36 97.46 97.41 e-vp 4768 4735 4666 98.54 97.86 98.20 i-vp 2602 2644 2527 95.57 97.12 96.34 e-adjp 384 372 332 89.25 86.46 87.83 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 818 724 88.51 88.08 88.29 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 492 466 94.72 92.64 93.67 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.39 74.62 75.98 Avg2. 46451 46451 45321 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11779 96.36 96.39 96.38 pp 4633 4661 4572 98.09 98.68 98.39 vp 4768 4735 4560 96.30 95.64 95.97 sbar 503 492 460 93.50 91.45 92.46 adjp 384 372 317 85.22 82.55 83.86 advp 822 818 716 87.53 87.10 87.32 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.40 80.60 81.49 Avg2. 23486 23457 22549 96.13 96.01 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 127 Log-likelihood = -39960.698763 Norm (log-likelihood gradient vector) = 951.136543 Norm (lambda vector) = 283.696712 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4583 98.28 98.92 98.60 i-np 13660 13720 13423 97.84 98.27 98.05 e-np 12220 12217 12033 98.49 98.47 98.48 o 6349 6346 6184 97.45 97.40 97.42 e-vp 4768 4732 4665 98.58 97.84 98.21 i-vp 2602 2644 2526 95.54 97.08 96.30 e-adjp 384 372 333 89.52 86.72 88.10 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 816 722 88.48 87.83 88.16 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 491 466 94.91 92.64 93.76 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 132 125 94.70 99.21 96.90 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.30 74.63 75.94 Avg2. 46451 46451 45330 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11785 96.46 96.44 96.45 pp 4633 4663 4573 98.07 98.70 98.39 vp 4768 4732 4557 96.30 95.57 95.94 sbar 503 491 460 93.69 91.45 92.56 adjp 384 372 317 85.22 82.55 83.86 advp 822 816 714 87.50 86.86 87.18 prt 126 132 125 94.70 99.21 96.90 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.42 80.58 81.49 Avg2. 23486 23446 22551 96.18 96.02 96.10 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 128 Log-likelihood = -39692.724498 Norm (log-likelihood gradient vector) = 1054.852799 Norm (lambda vector) = 284.370297 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4582 98.26 98.90 98.58 i-np 13660 13720 13426 97.86 98.29 98.07 e-np 12220 12218 12034 98.49 98.48 98.49 o 6349 6344 6184 97.48 97.40 97.44 e-vp 4768 4732 4666 98.61 97.86 98.23 i-vp 2602 2643 2526 95.57 97.08 96.32 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 817 721 88.25 87.71 87.98 i-advp 100 87 70 80.46 70.00 74.87 e-sbar 503 491 466 94.91 92.64 93.76 i-adjp 152 131 114 87.02 75.00 80.57 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.26 74.63 75.93 Avg2. 46451 46451 45334 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11786 96.46 96.45 96.46 pp 4633 4663 4572 98.05 98.68 98.36 vp 4768 4732 4558 96.32 95.60 95.96 sbar 503 491 460 93.69 91.45 92.56 adjp 384 372 318 85.48 82.81 84.13 advp 822 817 713 87.27 86.74 87.00 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.36 80.59 81.47 Avg2. 23486 23449 22552 96.17 96.02 96.10 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 129 Log-likelihood = -39459.244675 Norm (log-likelihood gradient vector) = 1101.655642 Norm (lambda vector) = 285.148248 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4649 4575 98.41 98.75 98.58 i-np 13660 14138 13522 95.64 98.99 97.29 e-np 12220 12019 11887 98.90 97.27 98.08 o 6349 6149 6035 98.15 95.05 96.58 e-vp 4768 4730 4662 98.56 97.78 98.17 i-vp 2602 2644 2527 95.57 97.12 96.34 e-adjp 384 370 332 89.73 86.46 88.06 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 810 717 88.52 87.23 87.87 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 497 469 94.37 93.24 93.80 i-adjp 152 129 112 86.82 73.68 79.72 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.29 73.88 75.55 Avg2. 46451 46451 45117 97.13 97.13 97.13 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12019 11553 96.12 94.54 95.33 pp 4633 4649 4565 98.19 98.53 98.36 vp 4768 4730 4555 96.30 95.53 95.91 sbar 503 497 463 93.16 92.05 92.60 adjp 384 370 315 85.14 82.03 83.55 advp 822 810 710 87.65 86.37 87.01 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 79.70 80.94 Avg2. 23486 23230 22305 96.02 94.97 95.49 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 130 Log-likelihood = -41848.873587 Norm (log-likelihood gradient vector) = 13420.743506 Norm (lambda vector) = 287.969821 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4660 4580 98.28 98.86 98.57 i-np 13660 13780 13441 97.54 98.40 97.97 e-np 12220 12191 12018 98.58 98.35 98.46 o 6349 6319 6169 97.63 97.16 97.40 e-vp 4768 4735 4669 98.61 97.92 98.26 i-vp 2602 2638 2526 95.75 97.08 96.41 e-adjp 384 371 334 90.03 86.98 88.48 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 815 720 88.34 87.59 87.97 i-advp 100 87 70 80.46 70.00 74.87 e-sbar 503 491 466 94.91 92.64 93.76 i-adjp 152 131 114 87.02 75.00 80.57 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.28 74.62 75.93 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12191 11760 96.46 96.24 96.35 pp 4633 4660 4570 98.07 98.64 98.35 vp 4768 4735 4564 96.39 95.72 96.05 sbar 503 491 460 93.69 91.45 92.56 adjp 384 371 318 85.71 82.81 84.24 advp 822 815 712 87.36 86.62 86.99 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.40 80.57 81.47 Avg2. 23486 23419 22529 96.20 95.93 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 131 Log-likelihood = -39308.660979 Norm (log-likelihood gradient vector) = 2807.128046 Norm (lambda vector) = 285.713501 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4659 4581 98.33 98.88 98.60 i-np 13660 13762 13435 97.62 98.35 97.99 e-np 12220 12200 12024 98.56 98.40 98.48 o 6349 6330 6174 97.54 97.24 97.39 e-vp 4768 4733 4667 98.61 97.88 98.24 i-vp 2602 2639 2525 95.68 97.04 96.36 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 813 716 88.07 87.10 87.58 i-advp 100 88 70 79.55 70.00 74.47 e-sbar 503 491 466 94.91 92.64 93.76 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.28 74.59 75.91 Avg2. 46451 46451 45317 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12200 11767 96.45 96.29 96.37 pp 4633 4659 4571 98.11 98.66 98.39 vp 4768 4733 4561 96.37 95.66 96.01 sbar 503 491 460 93.69 91.45 92.56 adjp 384 372 317 85.22 82.55 83.86 advp 822 813 708 87.08 86.13 86.61 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.32 80.50 81.40 Avg2. 23486 23424 22529 96.18 95.93 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 132 Log-likelihood = -39043.053010 Norm (log-likelihood gradient vector) = 1716.305232 Norm (lambda vector) = 286.609658 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4655 4578 98.35 98.81 98.58 i-np 13660 13721 13419 97.80 98.24 98.02 e-np 12220 12218 12035 98.50 98.49 98.49 o 6349 6356 6187 97.34 97.45 97.39 e-vp 4768 4737 4667 98.52 97.88 98.20 i-vp 2602 2637 2522 95.64 96.93 96.28 e-adjp 384 374 335 89.57 87.24 88.39 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 811 716 88.29 87.10 87.69 i-advp 100 86 70 81.40 70.00 75.27 e-sbar 503 494 467 94.53 92.84 93.68 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.31 74.10 75.67 Avg2. 46451 46451 45319 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11784 96.45 96.43 96.44 pp 4633 4655 4568 98.13 98.60 98.36 vp 4768 4737 4559 96.24 95.62 95.93 sbar 503 494 461 93.32 91.65 92.48 adjp 384 374 318 85.03 82.81 83.91 advp 822 811 709 87.42 86.25 86.83 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 79.93 81.06 Avg2. 23486 23444 22543 96.16 95.98 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 133 Log-likelihood = -38716.737150 Norm (log-likelihood gradient vector) = 869.811987 Norm (lambda vector) = 288.060493 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4656 4578 98.32 98.81 98.57 i-np 13660 13702 13410 97.87 98.17 98.02 e-np 12220 12223 12038 98.49 98.51 98.50 o 6349 6363 6192 97.31 97.53 97.42 e-vp 4768 4736 4666 98.52 97.86 98.19 i-vp 2602 2637 2522 95.64 96.93 96.28 e-adjp 384 375 335 89.33 87.24 88.27 i-pp 52 40 35 87.50 67.31 76.09 e-advp 822 818 720 88.02 87.59 87.80 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 494 467 94.53 92.84 93.68 i-adjp 152 132 114 86.36 75.00 80.28 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.33 74.12 75.69 Avg2. 46451 46451 45321 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11790 96.46 96.48 96.47 pp 4633 4656 4568 98.11 98.60 98.35 vp 4768 4736 4558 96.24 95.60 95.92 sbar 503 494 461 93.32 91.65 92.48 adjp 384 375 319 85.07 83.07 84.06 advp 822 818 713 87.16 86.74 86.95 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.20 80.01 81.09 Avg2. 23486 23457 22553 96.15 96.03 96.09 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 134 Log-likelihood = -38501.609305 Norm (log-likelihood gradient vector) = 993.937840 Norm (lambda vector) = 289.008153 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4658 4578 98.28 98.81 98.55 i-np 13660 13689 13408 97.95 98.16 98.05 e-np 12220 12223 12040 98.50 98.53 98.51 o 6349 6364 6195 97.34 97.57 97.46 e-vp 4768 4738 4666 98.48 97.86 98.17 i-vp 2602 2638 2522 95.60 96.93 96.26 e-adjp 384 376 336 89.36 87.50 88.42 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 822 723 87.96 87.96 87.96 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 494 467 94.53 92.84 93.68 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.28 74.22 75.72 Avg2. 46451 46451 45330 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11794 96.49 96.51 96.50 pp 4633 4658 4568 98.07 98.60 98.33 vp 4768 4738 4558 96.20 95.60 95.90 sbar 503 494 461 93.32 91.65 92.48 adjp 384 376 320 85.11 83.33 84.21 advp 822 822 716 87.10 87.10 87.10 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.19 80.08 81.12 Avg2. 23486 23466 22561 96.14 96.06 96.10 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 135 Log-likelihood = -38103.879445 Norm (log-likelihood gradient vector) = 1166.801771 Norm (lambda vector) = 290.625530 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4656 4578 98.32 98.81 98.57 i-np 13660 13637 13388 98.17 98.01 98.09 e-np 12220 12244 12049 98.41 98.60 98.50 o 6349 6385 6207 97.21 97.76 97.49 e-vp 4768 4743 4668 98.42 97.90 98.16 i-vp 2602 2642 2525 95.57 97.04 96.30 e-adjp 384 376 336 89.36 87.50 88.42 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 823 723 87.85 87.96 87.90 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 495 468 94.55 93.04 93.79 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.28 74.29 75.76 Avg2. 46451 46451 45338 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12244 11807 96.43 96.62 96.53 pp 4633 4656 4567 98.09 98.58 98.33 vp 4768 4743 4560 96.14 95.64 95.89 sbar 503 495 462 93.33 91.85 92.59 adjp 384 376 320 85.11 83.33 84.21 advp 822 823 716 87.00 87.10 87.05 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.18 80.11 81.13 Avg2. 23486 23492 22576 96.10 96.13 96.11 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 136 Log-likelihood = -37386.101940 Norm (log-likelihood gradient vector) = 1875.123532 Norm (lambda vector) = 293.963710 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4582 98.26 98.90 98.58 i-np 13660 13685 13408 97.98 98.16 98.07 e-np 12220 12227 12038 98.45 98.51 98.48 o 6349 6350 6188 97.45 97.46 97.46 e-vp 4768 4739 4666 98.46 97.86 98.16 i-vp 2602 2644 2527 95.57 97.12 96.34 e-adjp 384 373 335 89.81 87.24 88.51 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 824 725 87.99 88.20 88.09 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 490 466 95.10 92.64 93.86 i-adjp 152 138 116 84.06 76.32 80.00 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.26 74.79 76.00 Avg2. 46451 46451 45333 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11793 96.45 96.51 96.48 pp 4633 4663 4571 98.03 98.66 98.34 vp 4768 4739 4558 96.18 95.60 95.89 sbar 503 490 460 93.88 91.45 92.65 adjp 384 373 318 85.25 82.81 84.02 advp 822 824 718 87.14 87.35 87.24 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.32 80.66 81.48 Avg2. 23486 23472 22563 96.13 96.07 96.10 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 137 Log-likelihood = -36723.736361 Norm (log-likelihood gradient vector) = 1099.282937 Norm (lambda vector) = 296.664211 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4582 98.28 98.90 98.59 i-np 13660 13690 13410 97.95 98.17 98.06 e-np 12220 12225 12037 98.46 98.50 98.48 o 6349 6346 6185 97.46 97.42 97.44 e-vp 4768 4739 4667 98.48 97.88 98.18 i-vp 2602 2646 2528 95.54 97.16 96.34 e-adjp 384 373 336 90.08 87.50 88.77 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 822 723 87.96 87.96 87.96 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 492 468 95.12 93.04 94.07 i-adjp 152 138 117 84.78 76.97 80.69 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.21 74.80 75.99 Avg2. 46451 46451 45334 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12225 11791 96.45 96.49 96.47 pp 4633 4662 4571 98.05 98.66 98.35 vp 4768 4739 4559 96.20 95.62 95.91 sbar 503 492 462 93.90 91.85 92.86 adjp 384 373 319 85.52 83.07 84.28 advp 822 822 716 87.10 87.10 87.10 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.35 80.62 81.48 Avg2. 23486 23468 22562 96.14 96.07 96.10 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 138 Log-likelihood = -36400.604363 Norm (log-likelihood gradient vector) = 814.914885 Norm (lambda vector) = 296.914851 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4585 98.22 98.96 98.59 i-np 13660 13729 13422 97.76 98.26 98.01 e-np 12220 12213 12031 98.51 98.45 98.48 o 6349 6335 6181 97.57 97.35 97.46 e-vp 4768 4734 4663 98.50 97.80 98.15 i-vp 2602 2644 2525 95.50 97.04 96.26 e-adjp 384 371 335 90.30 87.24 88.74 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 816 721 88.36 87.71 88.03 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 488 465 95.29 92.45 93.84 i-adjp 152 136 116 85.29 76.32 80.56 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.61 75.07 76.32 Avg2. 46451 46451 45325 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11779 96.45 96.39 96.42 pp 4633 4668 4574 97.99 98.73 98.36 vp 4768 4734 4556 96.24 95.55 95.90 sbar 503 488 461 94.47 91.65 93.04 adjp 384 371 318 85.71 82.81 84.24 advp 822 816 714 87.50 86.86 87.18 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.46 80.54 81.49 Avg2. 23486 23445 22546 96.17 96.00 96.08 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 139 Log-likelihood = -35793.207794 Norm (log-likelihood gradient vector) = 1281.096408 Norm (lambda vector) = 298.136943 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4583 98.31 98.92 98.61 i-np 13660 13655 13391 98.07 98.03 98.05 e-np 12220 12246 12044 98.35 98.56 98.45 o 6349 6371 6196 97.25 97.59 97.42 e-vp 4768 4738 4664 98.44 97.82 98.13 i-vp 2602 2651 2528 95.36 97.16 96.25 e-adjp 384 367 332 90.46 86.46 88.42 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 496 468 94.35 93.04 93.69 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.34 75.06 76.18 Avg2. 46451 46451 45322 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12246 11793 96.30 96.51 96.40 pp 4633 4662 4572 98.07 98.68 98.38 vp 4768 4738 4558 96.20 95.60 95.90 sbar 503 496 463 93.35 92.05 92.69 adjp 384 367 316 86.10 82.29 84.15 advp 822 812 712 87.68 86.62 87.15 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.40 80.52 81.44 Avg2. 23486 23476 22558 96.09 96.05 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 140 Log-likelihood = -35618.679669 Norm (log-likelihood gradient vector) = 2238.132695 Norm (lambda vector) = 299.346465 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4582 98.31 98.90 98.60 i-np 13660 13705 13407 97.83 98.15 97.99 e-np 12220 12222 12032 98.45 98.46 98.45 o 6349 6350 6185 97.40 97.42 97.41 e-vp 4768 4739 4667 98.48 97.88 98.18 i-vp 2602 2646 2527 95.50 97.12 96.30 e-adjp 384 371 334 90.03 86.98 88.48 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 812 720 88.67 87.59 88.13 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 496 467 94.15 92.84 93.49 i-adjp 152 131 116 88.55 76.32 81.98 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.62 75.08 76.33 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12222 11778 96.37 96.38 96.38 pp 4633 4661 4571 98.07 98.66 98.36 vp 4768 4739 4561 96.24 95.66 95.95 sbar 503 496 463 93.35 92.05 92.69 adjp 384 371 318 85.71 82.81 84.24 advp 822 812 713 87.81 86.74 87.27 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.38 80.57 81.47 Avg2. 23486 23456 22548 96.13 96.01 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 141 Log-likelihood = -35227.251437 Norm (log-likelihood gradient vector) = 1047.478811 Norm (lambda vector) = 300.117124 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4657 4580 98.35 98.86 98.60 i-np 13660 13726 13415 97.73 98.21 97.97 e-np 12220 12213 12028 98.49 98.43 98.46 o 6349 6341 6179 97.45 97.32 97.38 e-vp 4768 4740 4667 98.46 97.88 98.17 i-vp 2602 2645 2526 95.50 97.08 96.28 e-adjp 384 374 336 89.84 87.50 88.65 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 495 468 94.55 93.04 93.79 i-adjp 152 131 115 87.79 75.66 81.27 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.68 75.07 76.35 Avg2. 46451 46451 45314 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11772 96.39 96.33 96.36 pp 4633 4657 4569 98.11 98.62 98.36 vp 4768 4740 4560 96.20 95.64 95.92 sbar 503 495 464 93.74 92.25 92.99 adjp 384 374 318 85.03 82.81 83.91 advp 822 812 712 87.68 86.62 87.15 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.34 80.57 81.44 Avg2. 23486 23446 22539 96.13 95.97 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 142 Log-likelihood = -34862.771918 Norm (log-likelihood gradient vector) = 1176.355822 Norm (lambda vector) = 301.521840 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4583 98.24 98.92 98.58 i-np 13660 13714 13413 97.81 98.19 98.00 e-np 12220 12218 12033 98.49 98.47 98.48 o 6349 6345 6182 97.43 97.37 97.40 e-vp 4768 4737 4665 98.48 97.84 98.16 i-vp 2602 2643 2524 95.50 97.00 96.24 e-adjp 384 375 336 89.60 87.50 88.54 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 816 721 88.36 87.71 88.03 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 490 464 94.69 92.25 93.45 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.49 74.52 75.98 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11780 96.42 96.40 96.41 pp 4633 4665 4572 98.01 98.68 98.34 vp 4768 4737 4558 96.22 95.60 95.91 sbar 503 490 460 93.88 91.45 92.65 adjp 384 375 318 84.80 82.81 83.79 advp 822 816 714 87.50 86.86 87.18 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.24 79.90 81.05 Avg2. 23486 23455 22545 96.12 95.99 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 143 Log-likelihood = -34468.877503 Norm (log-likelihood gradient vector) = 942.698424 Norm (lambda vector) = 303.244741 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4580 98.24 98.86 98.55 i-np 13660 13730 13416 97.71 98.21 97.96 e-np 12220 12209 12025 98.49 98.40 98.45 o 6349 6339 6177 97.44 97.29 97.37 e-vp 4768 4734 4663 98.50 97.80 98.15 i-vp 2602 2643 2524 95.50 97.00 96.24 e-adjp 384 377 336 89.12 87.50 88.30 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 816 722 88.48 87.83 88.16 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 494 466 94.33 92.64 93.48 i-adjp 152 132 116 87.88 76.32 81.69 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.49 74.53 75.98 Avg2. 46451 46451 45303 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12209 11769 96.40 96.31 96.35 pp 4633 4662 4569 98.01 98.62 98.31 vp 4768 4734 4555 96.22 95.53 95.87 sbar 503 494 462 93.52 91.85 92.68 adjp 384 377 320 84.88 83.33 84.10 advp 822 816 715 87.62 86.98 87.30 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 79.90 81.04 Avg2. 23486 23445 22532 96.11 95.94 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 144 Log-likelihood = -33905.585929 Norm (log-likelihood gradient vector) = 1346.920267 Norm (lambda vector) = 306.060759 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4675 4586 98.10 98.99 98.54 i-np 13660 13656 13388 98.04 98.01 98.02 e-np 12220 12246 12042 98.33 98.54 98.44 o 6349 6374 6196 97.21 97.59 97.40 e-vp 4768 4734 4663 98.50 97.80 98.15 i-vp 2602 2640 2524 95.61 97.00 96.30 e-adjp 384 378 336 88.89 87.50 88.19 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 820 724 88.29 88.08 88.19 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 479 456 95.20 90.66 92.87 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.53 74.51 75.99 Avg2. 46451 46451 45310 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12246 11789 96.27 96.47 96.37 pp 4633 4675 4575 97.86 98.75 98.30 vp 4768 4734 4557 96.26 95.57 95.92 sbar 503 479 452 94.36 89.86 92.06 adjp 384 378 319 84.39 83.07 83.73 advp 822 820 717 87.44 87.23 87.33 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.21 79.73 80.95 Avg2. 23486 23485 22551 96.02 96.02 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 145 Log-likelihood = -33289.192340 Norm (log-likelihood gradient vector) = 1997.163269 Norm (lambda vector) = 308.875133 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4585 98.14 98.96 98.55 i-np 13660 13689 13404 97.92 98.13 98.02 e-np 12220 12229 12035 98.41 98.49 98.45 o 6349 6357 6185 97.29 97.42 97.36 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2638 2523 95.64 96.96 96.30 e-adjp 384 377 336 89.12 87.50 88.30 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 822 724 88.08 88.08 88.08 i-advp 100 86 71 82.56 71.00 76.34 e-sbar 503 484 460 95.04 91.45 93.21 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.43 74.51 75.94 Avg2. 46451 46451 45310 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12229 11783 96.35 96.42 96.39 pp 4633 4672 4574 97.90 98.73 98.31 vp 4768 4734 4557 96.26 95.57 95.92 sbar 503 484 456 94.21 90.66 92.40 adjp 384 377 318 84.35 82.81 83.57 advp 822 822 717 87.23 87.23 87.23 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.19 79.78 80.96 Avg2. 23486 23471 22547 96.06 96.00 96.03 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 146 Log-likelihood = -32996.552309 Norm (log-likelihood gradient vector) = 1037.560673 Norm (lambda vector) = 308.434328 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4585 98.18 98.96 98.57 i-np 13660 13722 13414 97.76 98.20 97.98 e-np 12220 12213 12027 98.48 98.42 98.45 o 6349 6342 6179 97.43 97.32 97.38 e-vp 4768 4735 4665 98.52 97.84 98.18 i-vp 2602 2637 2523 95.68 96.96 96.32 e-adjp 384 375 336 89.60 87.50 88.54 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 821 724 88.19 88.08 88.13 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 486 462 95.06 91.85 93.43 i-adjp 152 134 118 88.06 77.63 82.52 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.50 74.57 76.01 Avg2. 46451 46451 45311 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12213 11774 96.41 96.35 96.38 pp 4633 4670 4574 97.94 98.73 98.33 vp 4768 4735 4558 96.26 95.60 95.93 sbar 503 486 458 94.24 91.05 92.62 adjp 384 375 319 85.07 83.07 84.06 advp 822 821 717 87.33 87.23 87.28 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.28 79.84 81.04 Avg2. 23486 23453 22542 96.12 95.98 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 147 Log-likelihood = -32679.042194 Norm (log-likelihood gradient vector) = 661.458831 Norm (lambda vector) = 308.349039 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4584 98.18 98.94 98.56 i-np 13660 13721 13416 97.78 98.21 97.99 e-np 12220 12217 12030 98.47 98.45 98.46 o 6349 6343 6180 97.43 97.34 97.38 e-vp 4768 4738 4667 98.50 97.88 98.19 i-vp 2602 2632 2522 95.82 96.93 96.37 e-adjp 384 372 335 90.05 87.24 88.62 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 818 722 88.26 87.83 88.05 i-advp 100 85 70 82.35 70.00 75.68 e-sbar 503 486 462 95.06 91.85 93.43 i-adjp 152 139 119 85.61 78.29 81.79 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.41 74.58 75.97 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11776 96.39 96.37 96.38 pp 4633 4669 4573 97.94 98.70 98.32 vp 4768 4738 4562 96.29 95.68 95.98 sbar 503 486 458 94.24 91.05 92.62 adjp 384 372 318 85.48 82.81 84.13 advp 822 818 715 87.41 86.98 87.20 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.33 79.80 81.04 Avg2. 23486 23453 22544 96.12 95.99 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 148 Log-likelihood = -32425.263661 Norm (log-likelihood gradient vector) = 709.862672 Norm (lambda vector) = 308.967204 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4580 98.26 98.86 98.56 i-np 13660 13816 13445 97.31 98.43 97.87 e-np 12220 12173 11999 98.57 98.19 98.38 o 6349 6295 6146 97.63 96.80 97.22 e-vp 4768 4741 4666 98.42 97.86 98.14 i-vp 2602 2637 2525 95.75 97.04 96.39 e-adjp 384 369 334 90.51 86.98 88.71 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 815 723 88.71 87.96 88.33 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 490 465 94.90 92.45 93.66 i-adjp 152 136 117 86.03 76.97 81.25 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.78 74.60 75.67 Avg2. 46451 46451 45280 97.48 97.48 97.48 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12173 11730 96.36 95.99 96.18 pp 4633 4661 4569 98.03 98.62 98.32 vp 4768 4741 4562 96.22 95.68 95.95 sbar 503 490 460 93.88 91.45 92.65 adjp 384 369 318 86.18 82.81 84.46 advp 822 815 717 87.98 87.23 87.60 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.79 79.89 80.83 Avg2. 23486 23403 22499 96.14 95.80 95.97 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 149 Log-likelihood = -32105.746466 Norm (log-likelihood gradient vector) = 5425.061866 Norm (lambda vector) = 313.826342 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4584 98.20 98.94 98.57 i-np 13660 13745 13432 97.72 98.33 98.03 e-np 12220 12208 12027 98.52 98.42 98.47 o 6349 6334 6178 97.54 97.31 97.42 e-vp 4768 4737 4666 98.50 97.86 98.18 i-vp 2602 2634 2522 95.75 96.93 96.33 e-adjp 384 370 335 90.54 87.24 88.86 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 815 722 88.59 87.83 88.21 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 487 463 95.07 92.05 93.54 i-adjp 152 137 118 86.13 77.63 81.66 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 11 78.57 91.67 84.62 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.54 74.60 76.04 Avg2. 46451 46451 45326 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12208 11777 96.47 96.37 96.42 pp 4633 4668 4573 97.96 98.70 98.33 vp 4768 4737 4560 96.26 95.64 95.95 sbar 503 487 459 94.25 91.25 92.73 adjp 384 370 318 85.95 82.81 84.35 advp 822 815 716 87.85 87.10 87.48 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.44 79.90 81.15 Avg2. 23486 23439 22546 96.19 96.00 96.09 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 150 Log-likelihood = -31995.999353 Norm (log-likelihood gradient vector) = 2040.633697 Norm (lambda vector) = 310.806383 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4659 4579 98.28 98.83 98.56 i-np 13660 13701 13414 97.91 98.20 98.05 e-np 12220 12224 12040 98.49 98.53 98.51 o 6349 6351 6185 97.39 97.42 97.40 e-vp 4768 4741 4666 98.42 97.86 98.14 i-vp 2602 2636 2524 95.75 97.00 96.37 e-adjp 384 374 337 90.11 87.76 88.92 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 814 721 88.57 87.71 88.14 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 491 466 94.91 92.64 93.76 i-adjp 152 141 120 85.11 78.95 81.91 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.68 74.77 75.71 Avg2. 46451 46451 45332 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11793 96.47 96.51 96.49 pp 4633 4659 4568 98.05 98.60 98.32 vp 4768 4741 4561 96.20 95.66 95.93 sbar 503 491 461 93.89 91.65 92.76 adjp 384 374 321 85.83 83.59 84.70 advp 822 814 715 87.84 86.98 87.41 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.68 80.02 80.84 Avg2. 23486 23458 22562 96.18 96.07 96.12 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 151 Log-likelihood = -31498.300914 Norm (log-likelihood gradient vector) = 903.499815 Norm (lambda vector) = 312.950622 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4660 4580 98.28 98.86 98.57 i-np 13660 13706 13417 97.89 98.22 98.06 e-np 12220 12223 12040 98.50 98.53 98.51 o 6349 6350 6186 97.42 97.43 97.42 e-vp 4768 4740 4665 98.42 97.84 98.13 i-vp 2602 2639 2525 95.68 97.04 96.36 e-adjp 384 374 337 90.11 87.76 88.92 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 814 722 88.70 87.83 88.26 i-advp 100 84 71 84.52 71.00 77.17 e-sbar 503 488 465 95.29 92.45 93.84 i-adjp 152 139 118 84.89 77.63 81.10 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.79 74.70 75.73 Avg2. 46451 46451 45335 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11795 96.50 96.52 96.51 pp 4633 4660 4569 98.05 98.62 98.33 vp 4768 4740 4560 96.20 95.64 95.92 sbar 503 488 460 94.26 91.45 92.84 adjp 384 374 321 85.83 83.59 84.70 advp 822 814 716 87.96 87.10 87.53 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.74 80.01 80.86 Avg2. 23486 23454 22564 96.21 96.07 96.14 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 152 Log-likelihood = -31201.455901 Norm (log-likelihood gradient vector) = 731.846914 Norm (lambda vector) = 314.540736 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4661 4580 98.26 98.86 98.56 i-np 13660 13681 13404 97.98 98.13 98.05 e-np 12220 12233 12045 98.46 98.57 98.52 o 6349 6363 6195 97.36 97.57 97.47 e-vp 4768 4742 4667 98.42 97.88 98.15 i-vp 2602 2637 2525 95.75 97.04 96.39 e-adjp 384 373 336 90.08 87.50 88.77 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 817 724 88.62 88.08 88.35 i-advp 100 85 71 83.53 71.00 76.76 e-sbar 503 488 465 95.29 92.45 93.84 i-adjp 152 137 117 85.40 76.97 80.97 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.76 74.68 75.71 Avg2. 46451 46451 45338 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12233 11797 96.44 96.54 96.49 pp 4633 4661 4569 98.03 98.62 98.32 vp 4768 4742 4562 96.20 95.68 95.94 sbar 503 488 460 94.26 91.45 92.84 adjp 384 373 320 85.79 83.33 84.54 advp 822 817 718 87.88 87.35 87.61 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.72 80.01 80.86 Avg2. 23486 23469 22569 96.17 96.10 96.13 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 153 Log-likelihood = -30752.808388 Norm (log-likelihood gradient vector) = 1188.892339 Norm (lambda vector) = 317.010020 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4582 98.24 98.90 98.57 i-np 13660 13697 13409 97.90 98.16 98.03 e-np 12220 12227 12041 98.48 98.54 98.51 o 6349 6357 6192 97.40 97.53 97.47 e-vp 4768 4740 4667 98.46 97.88 98.17 i-vp 2602 2639 2525 95.68 97.04 96.36 e-adjp 384 371 335 90.30 87.24 88.74 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 818 725 88.63 88.20 88.41 i-advp 100 84 70 83.33 70.00 76.09 e-sbar 503 487 464 95.28 92.25 93.74 i-adjp 152 136 116 85.29 76.32 80.56 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.43 74.58 75.97 Avg2. 46451 46451 45335 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11792 96.44 96.50 96.47 pp 4633 4664 4572 98.03 98.68 98.35 vp 4768 4740 4561 96.22 95.66 95.94 sbar 503 487 459 94.25 91.25 92.73 adjp 384 371 319 85.98 83.07 84.50 advp 822 818 719 87.90 87.47 87.68 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.44 79.98 81.19 Avg2. 23486 23461 22565 96.18 96.08 96.13 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 154 Log-likelihood = -30280.650695 Norm (log-likelihood gradient vector) = 864.105968 Norm (lambda vector) = 318.949826 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4584 98.22 98.94 98.58 i-np 13660 13699 13411 97.90 98.18 98.04 e-np 12220 12229 12043 98.48 98.55 98.52 o 6349 6356 6193 97.44 97.54 97.49 e-vp 4768 4738 4667 98.50 97.88 98.19 i-vp 2602 2644 2530 95.69 97.23 96.45 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 816 724 88.73 88.08 88.40 i-advp 100 83 70 84.34 70.00 76.50 e-sbar 503 485 464 95.67 92.25 93.93 i-adjp 152 130 112 86.15 73.68 79.43 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.42 74.44 75.90 Avg2. 46451 46451 45341 97.61 97.61 97.61 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12229 11795 96.45 96.52 96.49 pp 4633 4667 4574 98.01 98.73 98.37 vp 4768 4738 4562 96.29 95.68 95.98 sbar 503 485 459 94.64 91.25 92.91 adjp 384 372 317 85.22 82.55 83.86 advp 822 816 718 87.99 87.35 87.67 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.42 79.92 81.15 Avg2. 23486 23461 22568 96.19 96.09 96.14 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 155 Log-likelihood = -29775.479738 Norm (log-likelihood gradient vector) = 923.251761 Norm (lambda vector) = 320.913755 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4637 4566 98.47 98.55 98.51 i-np 13660 13777 13440 97.55 98.39 97.97 e-np 12220 12193 12018 98.56 98.35 98.46 o 6349 6320 6170 97.63 97.18 97.40 e-vp 4768 4735 4666 98.54 97.86 98.20 i-vp 2602 2648 2532 95.62 97.31 96.46 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 815 725 88.96 88.20 88.58 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 510 472 92.55 93.84 93.19 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.39 74.17 75.75 Avg2. 46451 46451 45316 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12193 11766 96.50 96.28 96.39 pp 4633 4637 4556 98.25 98.34 98.30 vp 4768 4735 4558 96.26 95.60 95.93 sbar 503 510 466 91.37 92.64 92.00 adjp 384 372 319 85.75 83.07 84.39 advp 822 815 719 88.22 87.47 87.84 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.20 80.06 81.11 Avg2. 23486 23416 22527 96.20 95.92 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 156 Log-likelihood = -29473.601752 Norm (log-likelihood gradient vector) = 3234.808908 Norm (lambda vector) = 325.461023 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4582 98.28 98.90 98.59 i-np 13660 13727 13430 97.84 98.32 98.08 e-np 12220 12218 12038 98.53 98.51 98.52 o 6349 6342 6189 97.59 97.48 97.53 e-vp 4768 4738 4667 98.50 97.88 98.19 i-vp 2602 2644 2531 95.73 97.27 96.49 e-adjp 384 371 334 90.03 86.98 88.48 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 817 725 88.74 88.20 88.47 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 489 466 95.30 92.64 93.95 i-adjp 152 129 112 86.82 73.68 79.72 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.51 74.47 75.96 Avg2. 46451 46451 45353 97.64 97.64 97.64 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11793 96.52 96.51 96.51 pp 4633 4662 4572 98.07 98.68 98.38 vp 4768 4738 4562 96.29 95.68 95.98 sbar 503 489 461 94.27 91.65 92.94 adjp 384 371 317 85.44 82.55 83.97 advp 822 817 719 88.00 87.47 87.74 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.42 79.97 81.18 Avg2. 23486 23449 22567 96.24 96.09 96.16 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 157 Log-likelihood = -29409.088424 Norm (log-likelihood gradient vector) = 1154.638926 Norm (lambda vector) = 322.755286 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4582 98.26 98.90 98.58 i-np 13660 13727 13431 97.84 98.32 98.08 e-np 12220 12218 12040 98.54 98.53 98.54 o 6349 6342 6188 97.57 97.46 97.52 e-vp 4768 4734 4665 98.54 97.84 98.19 i-vp 2602 2648 2531 95.58 97.27 96.42 e-adjp 384 371 333 89.76 86.72 88.21 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 815 724 88.83 88.08 88.45 i-advp 100 82 70 85.37 70.00 76.92 e-sbar 503 489 466 95.30 92.64 93.95 i-adjp 152 130 112 86.15 73.68 79.43 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 15 11 73.33 91.67 81.48 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.46 74.45 75.92 Avg2. 46451 46451 45351 97.63 97.63 97.63 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12218 11796 96.55 96.53 96.54 pp 4633 4663 4572 98.05 98.68 98.36 vp 4768 4734 4557 96.26 95.57 95.92 sbar 503 489 461 94.27 91.65 92.94 adjp 384 371 317 85.44 82.55 83.97 advp 822 815 718 88.10 87.35 87.72 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.43 79.95 81.17 Avg2. 23486 23444 22564 96.25 96.07 96.16 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 158 Log-likelihood = -29105.804973 Norm (log-likelihood gradient vector) = 870.008977 Norm (lambda vector) = 323.511121 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4582 98.28 98.90 98.59 i-np 13660 13703 13414 97.89 98.20 98.04 e-np 12220 12227 12040 98.47 98.53 98.50 o 6349 6349 6186 97.43 97.43 97.43 e-vp 4768 4734 4664 98.52 97.82 98.17 i-vp 2602 2650 2531 95.51 97.27 96.38 e-adjp 384 374 336 89.84 87.50 88.65 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 816 724 88.73 88.08 88.40 i-advp 100 81 70 86.42 70.00 77.35 e-sbar 503 490 466 95.10 92.64 93.86 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.53 74.22 75.84 Avg2. 46451 46451 45338 97.60 97.60 97.60 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12227 11795 96.47 96.52 96.49 pp 4633 4662 4572 98.07 98.68 98.38 vp 4768 4734 4555 96.22 95.53 95.87 sbar 503 490 460 93.88 91.45 92.65 adjp 384 374 321 85.83 83.59 84.70 advp 822 816 718 87.99 87.35 87.67 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.41 80.03 81.20 Avg2. 23486 23457 22564 96.19 96.07 96.13 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 159 Log-likelihood = -28646.207447 Norm (log-likelihood gradient vector) = 600.933464 Norm (lambda vector) = 325.454650 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4580 98.22 98.86 98.54 i-np 13660 13741 13430 97.74 98.32 98.03 e-np 12220 12209 12027 98.51 98.42 98.46 o 6349 6327 6172 97.55 97.21 97.38 e-vp 4768 4733 4661 98.48 97.76 98.12 i-vp 2602 2652 2531 95.44 97.27 96.35 e-adjp 384 373 335 89.81 87.24 88.51 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 818 725 88.63 88.20 88.41 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 490 464 94.69 92.25 93.45 i-adjp 152 134 117 87.31 76.97 81.82 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 21 20 95.24 83.33 88.89 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.42 73.81 75.57 Avg2. 46451 46451 45320 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12209 11772 96.42 96.33 96.38 pp 4633 4663 4570 98.01 98.64 98.32 vp 4768 4733 4553 96.20 95.49 95.84 sbar 503 490 457 93.27 90.85 92.04 adjp 384 373 320 85.79 83.33 84.54 advp 822 818 719 87.90 87.47 87.68 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.32 79.93 81.11 Avg2. 23486 23440 22534 96.13 95.95 96.04 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 160 Log-likelihood = -28108.212852 Norm (log-likelihood gradient vector) = 2099.502463 Norm (lambda vector) = 329.399816 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4580 98.16 98.86 98.51 i-np 13660 13678 13405 98.00 98.13 98.07 e-np 12220 12240 12045 98.41 98.57 98.49 o 6349 6354 6189 97.40 97.48 97.44 e-vp 4768 4735 4662 98.46 97.78 98.12 i-vp 2602 2652 2532 95.48 97.31 96.38 e-adjp 384 373 333 89.28 86.72 87.98 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 816 724 88.73 88.08 88.40 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 489 463 94.68 92.05 93.35 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.80 73.79 75.26 Avg2. 46451 46451 45328 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11798 96.39 96.55 96.47 pp 4633 4666 4569 97.92 98.62 98.27 vp 4768 4735 4556 96.22 95.55 95.89 sbar 503 489 456 93.25 90.66 91.94 adjp 384 373 318 85.25 82.81 84.02 advp 822 816 718 87.99 87.35 87.67 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.56 79.87 80.71 Avg2. 23486 23474 22558 96.10 96.05 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 161 Log-likelihood = -27759.768016 Norm (log-likelihood gradient vector) = 1038.678180 Norm (lambda vector) = 332.326914 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4578 98.18 98.81 98.49 i-np 13660 13689 13411 97.97 98.18 98.07 e-np 12220 12237 12045 98.43 98.57 98.50 o 6349 6350 6187 97.43 97.45 97.44 e-vp 4768 4735 4662 98.46 97.78 98.12 i-vp 2602 2652 2532 95.48 97.31 96.38 e-adjp 384 372 333 89.52 86.72 88.10 i-pp 52 41 35 85.37 67.31 75.27 e-advp 822 816 724 88.73 88.08 88.40 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 489 463 94.68 92.05 93.35 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.81 73.79 75.27 Avg2. 46451 46451 45330 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12237 11798 96.41 96.55 96.48 pp 4633 4663 4567 97.94 98.58 98.26 vp 4768 4735 4556 96.22 95.55 95.89 sbar 503 489 456 93.25 90.66 91.94 adjp 384 372 318 85.48 82.81 84.13 advp 822 816 718 87.99 87.35 87.67 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.59 79.87 80.72 Avg2. 23486 23467 22556 96.12 96.04 96.08 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 162 Log-likelihood = -27673.683135 Norm (log-likelihood gradient vector) = 756.341608 Norm (lambda vector) = 332.359398 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4659 4576 98.22 98.77 98.49 i-np 13660 13706 13413 97.86 98.19 98.03 e-np 12220 12230 12043 98.47 98.55 98.51 o 6349 6342 6184 97.51 97.40 97.45 e-vp 4768 4734 4661 98.46 97.76 98.11 i-vp 2602 2650 2529 95.43 97.19 96.31 e-adjp 384 372 333 89.52 86.72 88.10 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 817 722 88.37 87.83 88.10 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 494 465 94.13 92.45 93.28 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.63 73.72 75.15 Avg2. 46451 46451 45319 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12230 11791 96.41 96.49 96.45 pp 4633 4659 4565 97.98 98.53 98.26 vp 4768 4734 4552 96.16 95.47 95.81 sbar 503 494 458 92.71 91.05 91.88 adjp 384 372 318 85.48 82.81 84.13 advp 822 817 716 87.64 87.10 87.37 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.49 79.86 80.67 Avg2. 23486 23461 22543 96.09 95.98 96.04 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 163 Log-likelihood = -27415.636885 Norm (log-likelihood gradient vector) = 585.439757 Norm (lambda vector) = 334.405815 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4580 98.16 98.86 98.51 i-np 13660 13757 13431 97.63 98.32 97.98 e-np 12220 12206 12029 98.55 98.44 98.49 o 6349 6317 6168 97.64 97.15 97.39 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2649 2529 95.47 97.19 96.32 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 815 721 88.47 87.71 88.09 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 488 462 94.67 91.85 93.24 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.59 73.72 75.13 Avg2. 46451 46451 45309 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12206 11773 96.45 96.34 96.40 pp 4633 4666 4568 97.90 98.60 98.25 vp 4768 4732 4553 96.22 95.49 95.85 sbar 503 488 455 93.24 90.46 91.83 adjp 384 372 319 85.75 83.07 84.39 advp 822 815 715 87.73 86.98 87.35 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.58 79.81 80.69 Avg2. 23486 23434 22526 96.13 95.91 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 164 Log-likelihood = -27188.316482 Norm (log-likelihood gradient vector) = 1562.093778 Norm (lambda vector) = 337.430834 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4580 98.20 98.86 98.53 i-np 13660 13731 13423 97.76 98.27 98.01 e-np 12220 12219 12035 98.49 98.49 98.49 o 6349 6329 6177 97.60 97.29 97.44 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2649 2529 95.47 97.19 96.32 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 815 721 88.47 87.71 88.09 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 492 464 94.31 92.25 93.27 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.61 73.75 75.15 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12219 11784 96.44 96.43 96.44 pp 4633 4664 4568 97.94 98.60 98.27 vp 4768 4732 4553 96.22 95.49 95.85 sbar 503 492 457 92.89 90.85 91.86 adjp 384 372 319 85.75 83.07 84.39 advp 822 815 715 87.73 86.98 87.35 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.62 79.86 80.73 Avg2. 23486 23448 22539 96.12 95.97 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 165 Log-likelihood = -26914.182262 Norm (log-likelihood gradient vector) = 849.769205 Norm (lambda vector) = 339.971090 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4580 98.18 98.86 98.52 i-np 13660 13702 13414 97.90 98.20 98.05 e-np 12220 12230 12043 98.47 98.55 98.51 o 6349 6348 6190 97.51 97.50 97.50 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2653 2531 95.40 97.27 96.33 e-adjp 384 371 333 89.76 86.72 88.21 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 812 720 88.67 87.59 88.13 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 491 463 94.30 92.05 93.16 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.62 73.70 75.13 Avg2. 46451 46451 45328 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12230 11794 96.43 96.51 96.47 pp 4633 4665 4568 97.92 98.60 98.26 vp 4768 4732 4555 96.26 95.53 95.89 sbar 503 491 456 92.87 90.66 91.75 adjp 384 371 318 85.71 82.81 84.24 advp 822 812 714 87.93 86.86 87.39 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.57 79.81 80.68 Avg2. 23486 23456 22548 96.13 96.01 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 166 Log-likelihood = -26633.010562 Norm (log-likelihood gradient vector) = 674.722726 Norm (lambda vector) = 342.900178 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4581 98.14 98.88 98.51 i-np 13660 13698 13412 97.91 98.18 98.05 e-np 12220 12232 12044 98.46 98.56 98.51 o 6349 6350 6191 97.50 97.51 97.50 e-vp 4768 4731 4660 98.50 97.73 98.12 i-vp 2602 2652 2531 95.44 97.27 96.35 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 812 721 88.79 87.71 88.25 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 487 461 94.66 91.65 93.13 i-adjp 152 134 117 87.31 76.97 81.82 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.72 73.81 75.24 Avg2. 46451 46451 45331 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12232 11796 96.44 96.53 96.48 pp 4633 4668 4569 97.88 98.62 98.25 vp 4768 4731 4554 96.26 95.51 95.88 sbar 503 487 454 93.22 90.26 91.72 adjp 384 372 319 85.75 83.07 84.39 advp 822 812 715 88.05 86.98 87.52 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.62 79.81 80.70 Avg2. 23486 23457 22550 96.13 96.01 96.07 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 167 Log-likelihood = -26414.863470 Norm (log-likelihood gradient vector) = 902.230948 Norm (lambda vector) = 345.340586 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4580 98.14 98.86 98.49 i-np 13660 13691 13412 97.96 98.18 98.07 e-np 12220 12236 12043 98.42 98.55 98.49 o 6349 6354 6189 97.40 97.48 97.44 e-vp 4768 4732 4660 98.48 97.73 98.11 i-vp 2602 2653 2531 95.40 97.27 96.33 e-adjp 384 373 334 89.54 86.98 88.24 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 809 719 88.88 87.47 88.17 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 487 461 94.66 91.65 93.13 i-adjp 152 134 117 87.31 76.97 81.82 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.70 73.80 75.22 Avg2. 46451 46451 45325 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12236 11796 96.40 96.53 96.47 pp 4633 4667 4568 97.88 98.60 98.24 vp 4768 4732 4553 96.22 95.49 95.85 sbar 503 487 454 93.22 90.26 91.72 adjp 384 373 319 85.52 83.07 84.28 advp 822 809 713 88.13 86.74 87.43 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.59 79.79 80.68 Avg2. 23486 23459 22546 96.11 96.00 96.05 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 168 Log-likelihood = -26088.376316 Norm (log-likelihood gradient vector) = 727.133593 Norm (lambda vector) = 349.536656 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4581 98.12 98.88 98.49 i-np 13660 13637 13375 98.08 97.91 98.00 e-np 12220 12256 12050 98.32 98.61 98.46 o 6349 6379 6199 97.18 97.64 97.41 e-vp 4768 4733 4659 98.44 97.71 98.07 i-vp 2602 2657 2533 95.33 97.35 96.33 e-adjp 384 374 332 88.77 86.46 87.60 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 809 719 88.88 87.47 88.17 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 485 460 94.85 91.45 93.12 i-adjp 152 137 117 85.40 76.97 80.97 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.56 73.76 75.14 Avg2. 46451 46451 45304 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12256 11795 96.24 96.52 96.38 pp 4633 4669 4569 97.86 98.62 98.24 vp 4768 4733 4554 96.22 95.51 95.86 sbar 503 485 453 93.40 90.06 91.70 adjp 384 374 317 84.76 82.55 83.64 advp 822 809 713 88.13 86.74 87.43 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.52 79.72 80.61 Avg2. 23486 23481 22544 96.01 95.99 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 169 Log-likelihood = -26204.460887 Norm (log-likelihood gradient vector) = 3220.411148 Norm (lambda vector) = 353.448082 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4579 98.14 98.83 98.48 i-np 13660 13675 13401 98.00 98.10 98.05 e-np 12220 12240 12046 98.42 98.58 98.50 o 6349 6362 6192 97.33 97.53 97.43 e-vp 4768 4732 4660 98.48 97.73 98.11 i-vp 2602 2655 2533 95.40 97.35 96.37 e-adjp 384 375 334 89.07 86.98 88.01 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 808 719 88.99 87.47 88.22 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 487 461 94.66 91.65 93.13 i-adjp 152 136 117 86.03 76.97 81.25 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.62 73.80 75.18 Avg2. 46451 46451 45321 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11799 96.40 96.55 96.48 pp 4633 4666 4567 97.88 98.58 98.23 vp 4768 4732 4555 96.26 95.53 95.89 sbar 503 487 454 93.22 90.26 91.72 adjp 384 375 319 85.07 83.07 84.06 advp 822 808 713 88.24 86.74 87.48 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.56 79.79 80.67 Avg2. 23486 23463 22550 96.11 96.01 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 170 Log-likelihood = -26019.319960 Norm (log-likelihood gradient vector) = 1288.767121 Norm (lambda vector) = 350.936222 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4581 98.14 98.88 98.51 i-np 13660 13693 13405 97.90 98.13 98.01 e-np 12220 12231 12042 98.45 98.54 98.50 o 6349 6353 6189 97.42 97.48 97.45 e-vp 4768 4733 4661 98.48 97.76 98.12 i-vp 2602 2656 2533 95.37 97.35 96.35 e-adjp 384 373 332 89.01 86.46 87.71 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 807 718 88.97 87.35 88.15 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 486 461 94.86 91.65 93.23 i-adjp 152 136 117 86.03 76.97 81.25 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.63 73.77 75.17 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11794 96.43 96.51 96.47 pp 4633 4668 4569 97.88 98.62 98.25 vp 4768 4733 4555 96.24 95.53 95.88 sbar 503 486 454 93.42 90.26 91.81 adjp 384 373 317 84.99 82.55 83.75 advp 822 807 712 88.23 86.62 87.42 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.57 79.73 80.64 Avg2. 23486 23453 22544 96.12 95.99 96.06 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 171 Log-likelihood = -25854.535116 Norm (log-likelihood gradient vector) = 890.006188 Norm (lambda vector) = 352.260148 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4580 98.14 98.86 98.49 i-np 13660 13698 13405 97.86 98.13 98.00 e-np 12220 12230 12041 98.45 98.54 98.49 o 6349 6345 6182 97.43 97.37 97.40 e-vp 4768 4733 4660 98.46 97.73 98.09 i-vp 2602 2655 2533 95.40 97.35 96.37 e-adjp 384 375 334 89.07 86.98 88.01 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 807 719 89.10 87.47 88.28 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 487 461 94.66 91.65 93.13 i-adjp 152 139 118 84.89 77.63 81.10 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.57 73.83 75.17 Avg2. 46451 46451 45312 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12230 11786 96.37 96.45 96.41 pp 4633 4667 4568 97.88 98.60 98.24 vp 4768 4733 4555 96.24 95.53 95.88 sbar 503 487 454 93.22 90.26 91.72 adjp 384 375 317 84.53 82.55 83.53 advp 822 807 713 88.35 86.74 87.54 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.52 79.73 80.61 Avg2. 23486 23454 22536 96.09 95.96 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 172 Log-likelihood = -25697.903106 Norm (log-likelihood gradient vector) = 632.715894 Norm (lambda vector) = 353.459245 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4579 98.18 98.83 98.50 i-np 13660 13710 13412 97.83 98.18 98.01 e-np 12220 12226 12039 98.47 98.52 98.49 o 6349 6342 6181 97.46 97.35 97.41 e-vp 4768 4735 4662 98.46 97.78 98.12 i-vp 2602 2654 2533 95.44 97.35 96.39 e-adjp 384 374 333 89.04 86.72 87.86 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 807 719 89.10 87.47 88.28 i-advp 100 82 72 87.80 72.00 79.12 e-sbar 503 489 462 94.48 91.85 93.15 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.61 73.76 75.16 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12226 11784 96.38 96.43 96.41 pp 4633 4664 4567 97.92 98.58 98.25 vp 4768 4735 4557 96.24 95.57 95.91 sbar 503 489 455 93.05 90.46 91.73 adjp 384 374 317 84.76 82.55 83.64 advp 822 807 713 88.35 86.74 87.54 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.53 79.75 80.63 Avg2. 23486 23450 22536 96.10 95.96 96.03 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 173 Log-likelihood = -25516.471957 Norm (log-likelihood gradient vector) = 784.488094 Norm (lambda vector) = 355.273010 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4579 98.16 98.83 98.49 i-np 13660 13699 13406 97.86 98.14 98.00 e-np 12220 12234 12042 98.43 98.54 98.49 o 6349 6347 6183 97.42 97.39 97.40 e-vp 4768 4735 4663 98.48 97.80 98.14 i-vp 2602 2651 2530 95.44 97.23 96.33 e-adjp 384 374 334 89.30 86.98 88.13 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 808 717 88.74 87.23 87.98 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 489 462 94.48 91.85 93.15 i-adjp 152 134 115 85.82 75.66 80.42 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.50 73.67 75.06 Avg2. 46451 46451 45309 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11786 96.34 96.45 96.39 pp 4633 4665 4567 97.90 98.58 98.24 vp 4768 4735 4555 96.20 95.53 95.86 sbar 503 489 455 93.05 90.46 91.73 adjp 384 374 317 84.76 82.55 83.64 advp 822 808 711 88.00 86.50 87.24 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.48 79.72 80.59 Avg2. 23486 23460 22534 96.05 95.95 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 174 Log-likelihood = -25294.139621 Norm (log-likelihood gradient vector) = 790.433894 Norm (lambda vector) = 357.982069 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4672 4582 98.07 98.90 98.48 i-np 13660 13795 13444 97.46 98.42 97.93 e-np 12220 12194 12014 98.52 98.31 98.42 o 6349 6295 6152 97.73 96.90 97.31 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2647 2528 95.50 97.16 96.32 e-adjp 384 374 336 89.84 87.50 88.65 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 808 719 88.99 87.47 88.22 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 486 459 94.44 91.25 92.82 i-adjp 152 132 115 87.12 75.66 80.99 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.49 73.66 75.04 Avg2. 46451 46451 45288 97.50 97.50 97.50 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12194 11750 96.36 96.15 96.26 pp 4633 4672 4568 97.77 98.60 98.18 vp 4768 4732 4552 96.20 95.47 95.83 sbar 503 486 452 93.00 89.86 91.41 adjp 384 374 318 85.03 82.81 83.91 advp 822 808 713 88.24 86.74 87.48 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.52 79.68 80.59 Avg2. 23486 23421 22496 96.05 95.78 95.92 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 175 Log-likelihood = -25257.175046 Norm (log-likelihood gradient vector) = 2342.757976 Norm (lambda vector) = 362.358206 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4581 98.12 98.88 98.49 i-np 13660 13714 13409 97.78 98.16 97.97 e-np 12220 12232 12036 98.40 98.49 98.45 o 6349 6336 6175 97.46 97.26 97.36 e-vp 4768 4736 4664 98.48 97.82 98.15 i-vp 2602 2648 2530 95.54 97.23 96.38 e-adjp 384 373 335 89.81 87.24 88.51 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 807 717 88.85 87.23 88.03 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 487 460 94.46 91.45 92.93 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 20 90.91 83.33 86.96 e-conjp 16 13 11 84.62 68.75 75.86 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.48 73.69 75.06 Avg2. 46451 46451 45301 97.52 97.52 97.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12232 11774 96.26 96.35 96.30 pp 4633 4669 4568 97.84 98.60 98.22 vp 4768 4736 4557 96.22 95.57 95.90 sbar 503 487 453 93.02 90.06 91.52 adjp 384 373 318 85.25 82.81 84.02 advp 822 807 711 88.10 86.50 87.29 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 11 84.62 68.75 75.86 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.52 79.71 80.60 Avg2. 23486 23459 22524 96.01 95.90 95.96 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 176 Log-likelihood = -24889.884985 Norm (log-likelihood gradient vector) = 675.213475 Norm (lambda vector) = 362.911530 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4581 98.12 98.88 98.49 i-np 13660 13697 13406 97.88 98.14 98.01 e-np 12220 12238 12042 98.40 98.54 98.47 o 6349 6346 6184 97.45 97.40 97.42 e-vp 4768 4735 4663 98.48 97.80 98.14 i-vp 2602 2650 2531 95.51 97.27 96.38 e-adjp 384 372 334 89.78 86.98 88.36 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 808 718 88.86 87.35 88.10 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 487 460 94.46 91.45 92.93 i-adjp 152 134 117 87.31 76.97 81.82 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 23 21 91.30 87.50 89.36 e-conjp 16 14 12 85.71 75.00 80.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.73 74.21 75.45 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12238 11786 96.31 96.45 96.38 pp 4633 4669 4569 97.86 98.62 98.24 vp 4768 4735 4556 96.22 95.55 95.89 sbar 503 487 453 93.02 90.06 91.52 adjp 384 372 318 85.48 82.81 84.13 advp 822 808 712 88.12 86.62 87.36 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 12 85.71 75.00 80.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.66 80.27 80.96 Avg2. 23486 23464 22537 96.05 95.96 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 177 Log-likelihood = -24745.157803 Norm (log-likelihood gradient vector) = 564.807457 Norm (lambda vector) = 363.148029 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4582 98.12 98.90 98.51 i-np 13660 13677 13401 97.98 98.10 98.04 e-np 12220 12246 12046 98.37 98.58 98.47 o 6349 6357 6188 97.34 97.46 97.40 e-vp 4768 4738 4663 98.42 97.80 98.11 i-vp 2602 2645 2527 95.54 97.12 96.32 e-adjp 384 374 334 89.30 86.98 88.13 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 808 718 88.86 87.35 88.10 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 486 460 94.65 91.45 93.02 i-adjp 152 134 117 87.31 76.97 81.82 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 21 91.30 87.50 89.36 e-conjp 16 14 12 85.71 75.00 80.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.86 74.62 75.73 Avg2. 46451 46451 45316 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12246 11791 96.28 96.49 96.39 pp 4633 4670 4571 97.88 98.66 98.27 vp 4768 4738 4557 96.18 95.57 95.88 sbar 503 486 454 93.42 90.26 91.81 adjp 384 374 318 85.03 82.81 83.91 advp 822 808 712 88.12 86.62 87.36 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 12 85.71 75.00 80.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.65 80.30 80.97 Avg2. 23486 23477 22546 96.03 96.00 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 178 Log-likelihood = -24570.256390 Norm (log-likelihood gradient vector) = 806.501789 Norm (lambda vector) = 364.185543 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4582 98.16 98.90 98.53 i-np 13660 13692 13406 97.91 98.14 98.03 e-np 12220 12234 12038 98.40 98.51 98.45 o 6349 6353 6186 97.37 97.43 97.40 e-vp 4768 4736 4660 98.40 97.73 98.06 i-vp 2602 2643 2525 95.54 97.04 96.28 e-adjp 384 378 336 88.89 87.50 88.19 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 485 459 94.64 91.25 92.91 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 23 21 91.30 87.50 89.36 e-conjp 16 14 12 85.71 75.00 80.00 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.89 74.64 75.75 Avg2. 46451 46451 45308 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11782 96.31 96.42 96.36 pp 4633 4668 4571 97.92 98.66 98.29 vp 4768 4736 4552 96.11 95.47 95.79 sbar 503 485 453 93.40 90.06 91.70 adjp 384 378 320 84.66 83.33 83.99 advp 822 812 713 87.81 86.74 87.27 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 14 12 85.71 75.00 80.00 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.58 80.33 80.95 Avg2. 23486 23468 22534 96.02 95.95 95.98 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 179 Log-likelihood = -24303.582175 Norm (log-likelihood gradient vector) = 781.617945 Norm (lambda vector) = 365.911640 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4579 98.09 98.83 98.46 i-np 13660 13673 13393 97.95 98.05 98.00 e-np 12220 12241 12041 98.37 98.54 98.45 o 6349 6359 6190 97.34 97.50 97.42 e-vp 4768 4734 4659 98.42 97.71 98.06 i-vp 2602 2644 2525 95.50 97.04 96.26 e-adjp 384 379 336 88.65 87.50 88.07 i-pp 52 42 35 83.33 67.31 74.47 e-advp 822 817 720 88.13 87.59 87.86 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 485 457 94.23 90.85 92.51 i-adjp 152 136 118 86.76 77.63 81.94 e-prt 126 131 123 93.89 97.62 95.72 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.27 74.65 75.94 Avg2. 46451 46451 45298 97.52 97.52 97.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12241 11782 96.25 96.42 96.33 pp 4633 4668 4567 97.84 98.58 98.20 vp 4768 4734 4551 96.13 95.45 95.79 sbar 503 485 451 92.99 89.66 91.30 adjp 384 379 321 84.70 83.59 84.14 advp 822 817 714 87.39 86.86 87.13 prt 126 131 123 93.89 97.62 95.72 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.15 80.32 81.22 Avg2. 23486 23478 22529 95.96 95.93 95.94 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 180 Log-likelihood = -24060.260080 Norm (log-likelihood gradient vector) = 1394.279069 Norm (lambda vector) = 369.753501 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4580 98.20 98.86 98.53 i-np 13660 13708 13406 97.80 98.14 97.97 e-np 12220 12225 12031 98.41 98.45 98.43 o 6349 6341 6178 97.43 97.31 97.37 e-vp 4768 4733 4659 98.44 97.71 98.07 i-vp 2602 2646 2527 95.50 97.12 96.30 e-adjp 384 378 336 88.89 87.50 88.19 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 816 720 88.24 87.59 87.91 i-advp 100 83 71 85.54 71.00 77.60 e-sbar 503 487 460 94.46 91.45 92.93 i-adjp 152 136 118 86.76 77.63 81.94 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.21 74.72 75.94 Avg2. 46451 46451 45296 97.51 97.51 97.51 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12225 11770 96.28 96.32 96.30 pp 4633 4664 4567 97.92 98.58 98.25 vp 4768 4733 4551 96.15 95.45 95.80 sbar 503 487 454 93.22 90.26 91.72 adjp 384 378 321 84.92 83.59 84.25 advp 822 816 714 87.50 86.86 87.18 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.22 80.45 81.33 Avg2. 23486 23458 22521 96.01 95.89 95.95 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 181 Log-likelihood = -23831.366179 Norm (log-likelihood gradient vector) = 593.431895 Norm (lambda vector) = 370.513765 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4580 98.24 98.86 98.55 i-np 13660 13718 13414 97.78 98.20 97.99 e-np 12220 12224 12034 98.45 98.48 98.46 o 6349 6334 6178 97.54 97.31 97.42 e-vp 4768 4732 4660 98.48 97.73 98.11 i-vp 2602 2652 2530 95.40 97.23 96.31 e-adjp 384 375 336 89.60 87.50 88.54 i-pp 52 44 36 81.82 69.23 75.00 e-advp 822 814 717 88.08 87.23 87.65 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 488 461 94.47 91.65 93.04 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 14 10 71.43 83.33 76.92 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.31 74.78 76.03 Avg2. 46451 46451 45309 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12224 11779 96.36 96.39 96.38 pp 4633 4662 4568 97.98 98.60 98.29 vp 4768 4732 4550 96.15 95.43 95.79 sbar 503 488 455 93.24 90.46 91.83 adjp 384 375 320 85.33 83.33 84.32 advp 822 814 711 87.35 86.50 86.92 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.27 80.41 81.33 Avg2. 23486 23450 22527 96.06 95.92 95.99 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 182 Log-likelihood = -23661.598927 Norm (log-likelihood gradient vector) = 634.685530 Norm (lambda vector) = 371.918639 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4580 98.20 98.86 98.53 i-np 13660 13709 13414 97.85 98.20 98.02 e-np 12220 12228 12039 98.45 98.52 98.49 o 6349 6335 6178 97.52 97.31 97.41 e-vp 4768 4734 4661 98.46 97.76 98.11 i-vp 2602 2648 2527 95.43 97.12 96.27 e-adjp 384 376 337 89.63 87.76 88.68 i-pp 52 46 36 78.26 69.23 73.47 e-advp 822 815 717 87.98 87.23 87.60 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 488 461 94.47 91.65 93.04 i-adjp 152 135 117 86.67 76.97 81.53 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 77.02 74.42 75.70 Avg2. 46451 46451 45313 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12228 11785 96.38 96.44 96.41 pp 4633 4664 4568 97.94 98.60 98.27 vp 4768 4734 4551 96.13 95.45 95.79 sbar 503 488 454 93.03 90.26 91.62 adjp 384 376 321 85.37 83.59 84.47 advp 822 815 711 87.24 86.50 86.87 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.24 80.50 81.36 Avg2. 23486 23461 22535 96.05 95.95 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 183 Log-likelihood = -23505.944033 Norm (log-likelihood gradient vector) = 661.562571 Norm (lambda vector) = 373.175011 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4673 4581 98.03 98.88 98.45 i-np 13660 13736 13421 97.71 98.25 97.98 e-np 12220 12215 12030 98.49 98.45 98.47 o 6349 6323 6169 97.56 97.16 97.36 e-vp 4768 4733 4662 98.50 97.78 98.14 i-vp 2602 2645 2526 95.50 97.08 96.28 e-adjp 384 379 338 89.18 88.02 88.60 i-pp 52 47 36 76.60 69.23 72.73 e-advp 822 811 717 88.41 87.23 87.81 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 482 456 94.61 90.66 92.59 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 0 0 0.00 0.00 0.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 76.94 74.34 75.61 Avg2. 46451 46451 45298 97.52 97.52 97.52 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12215 11773 96.38 96.34 96.36 pp 4633 4673 4569 97.77 98.62 98.19 vp 4768 4733 4550 96.13 95.43 95.78 sbar 503 482 449 93.15 89.26 91.17 adjp 384 379 320 84.43 83.33 83.88 advp 822 811 711 87.67 86.50 87.08 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 0 0 0.00 0.00 0.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.18 80.37 81.27 Avg2. 23486 23449 22517 96.03 95.87 95.95 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 184 Log-likelihood = -22990.285889 Norm (log-likelihood gradient vector) = 1340.967855 Norm (lambda vector) = 378.793635 Log-likelihood and gradient computational time: 322 seconds Training iteration elapsed: 323 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4659 4578 98.26 98.81 98.54 i-np 13660 13646 13383 98.07 97.97 98.02 e-np 12220 12257 12051 98.32 98.62 98.47 o 6349 6374 6195 97.19 97.57 97.38 e-vp 4768 4735 4664 98.50 97.82 98.16 i-vp 2602 2644 2529 95.65 97.19 96.42 e-adjp 384 380 338 88.95 88.02 88.48 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 809 717 88.63 87.23 87.92 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 494 466 94.33 92.64 93.48 i-adjp 152 134 116 86.57 76.32 81.12 e-prt 126 134 125 93.28 99.21 96.15 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.04 75.61 78.70 Avg2. 46451 46451 45319 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12257 11796 96.24 96.53 96.38 pp 4633 4659 4565 97.98 98.53 98.26 vp 4768 4735 4556 96.22 95.55 95.89 sbar 503 494 459 92.91 91.25 92.08 adjp 384 380 320 84.21 83.33 83.77 advp 822 809 711 87.89 86.50 87.19 prt 126 134 125 93.28 99.21 96.15 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.10 83.09 87.37 Avg2. 23486 23492 22553 96.00 96.03 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 357 seconds Iteration: 185 Log-likelihood = -22719.823989 Norm (log-likelihood gradient vector) = 1633.328315 Norm (lambda vector) = 382.963486 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4579 98.20 98.83 98.52 i-np 13660 13692 13408 97.93 98.16 98.04 e-np 12220 12235 12041 98.41 98.54 98.47 o 6349 6347 6183 97.42 97.39 97.40 e-vp 4768 4735 4664 98.50 97.82 98.16 i-vp 2602 2644 2529 95.65 97.19 96.42 e-adjp 384 379 338 89.18 88.02 88.60 i-pp 52 45 35 77.78 67.31 72.16 e-advp 822 812 719 88.55 87.47 88.00 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 491 464 94.50 92.25 93.36 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 134 125 93.28 99.21 96.15 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.01 75.60 78.67 Avg2. 46451 46451 45323 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12235 11784 96.31 96.43 96.37 pp 4633 4663 4566 97.92 98.55 98.24 vp 4768 4735 4556 96.22 95.55 95.89 sbar 503 491 457 93.08 90.85 91.95 adjp 384 379 321 84.70 83.59 84.14 advp 822 812 713 87.81 86.74 87.27 prt 126 134 125 93.28 99.21 96.15 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.16 83.09 87.39 Avg2. 23486 23473 22543 96.04 95.98 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 186 Log-likelihood = -22551.848808 Norm (log-likelihood gradient vector) = 555.921778 Norm (lambda vector) = 381.822146 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4665 4580 98.18 98.86 98.52 i-np 13660 13694 13412 97.94 98.18 98.06 e-np 12220 12233 12042 98.44 98.54 98.49 o 6349 6349 6187 97.45 97.45 97.45 e-vp 4768 4734 4663 98.50 97.80 98.15 i-vp 2602 2645 2529 95.61 97.19 96.40 e-adjp 384 381 339 88.98 88.28 88.63 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 810 718 88.64 87.35 87.99 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 488 462 94.67 91.85 93.24 i-adjp 152 133 116 87.22 76.32 81.40 e-prt 126 134 125 93.28 99.21 96.15 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.10 75.59 78.71 Avg2. 46451 46451 45330 97.59 97.59 97.59 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12233 11787 96.35 96.46 96.41 pp 4633 4665 4567 97.90 98.58 98.24 vp 4768 4734 4555 96.22 95.53 95.87 sbar 503 488 455 93.24 90.46 91.83 adjp 384 381 322 84.51 83.85 84.18 advp 822 810 712 87.90 86.62 87.25 prt 126 134 125 93.28 99.21 96.15 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.17 83.07 87.38 Avg2. 23486 23469 22544 96.06 95.99 96.02 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 187 Log-likelihood = -22451.640999 Norm (log-likelihood gradient vector) = 450.968925 Norm (lambda vector) = 381.953246 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4580 98.11 98.86 98.48 i-np 13660 13701 13415 97.91 98.21 98.06 e-np 12220 12228 12039 98.45 98.52 98.49 o 6349 6344 6183 97.46 97.39 97.42 e-vp 4768 4734 4662 98.48 97.78 98.13 i-vp 2602 2645 2530 95.65 97.23 96.44 e-adjp 384 383 340 88.77 88.54 88.66 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 810 718 88.64 87.35 87.99 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 485 459 94.64 91.25 92.91 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.06 75.57 78.68 Avg2. 46451 46451 45324 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12228 11784 96.37 96.43 96.40 pp 4633 4668 4567 97.84 98.58 98.20 vp 4768 4734 4556 96.24 95.55 95.90 sbar 503 485 452 93.20 89.86 91.50 adjp 384 383 322 84.07 83.85 83.96 advp 822 810 712 87.90 86.62 87.25 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.19 83.01 87.36 Avg2. 23486 23465 22539 96.05 95.97 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 188 Log-likelihood = -22191.586334 Norm (log-likelihood gradient vector) = 572.918804 Norm (lambda vector) = 383.660162 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4581 98.16 98.88 98.52 i-np 13660 13729 13419 97.74 98.24 97.99 e-np 12220 12217 12030 98.47 98.45 98.46 o 6349 6329 6176 97.58 97.28 97.43 e-vp 4768 4732 4661 98.50 97.76 98.13 i-vp 2602 2647 2531 95.62 97.27 96.44 e-adjp 384 382 339 88.74 88.28 88.51 i-pp 52 43 35 81.40 67.31 73.68 e-advp 822 810 717 88.52 87.23 87.87 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 487 461 94.66 91.65 93.13 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 133 125 93.98 99.21 96.53 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.20 75.57 78.74 Avg2. 46451 46451 45313 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12217 11772 96.36 96.33 96.35 pp 4633 4667 4569 97.90 98.62 98.26 vp 4768 4732 4554 96.24 95.51 95.87 sbar 503 487 454 93.22 90.26 91.72 adjp 384 382 321 84.03 83.59 83.81 advp 822 810 711 87.78 86.50 87.13 prt 126 133 125 93.98 99.21 96.53 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.18 83.00 87.35 Avg2. 23486 23452 22527 96.06 95.92 95.99 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 189 Log-likelihood = -21826.334435 Norm (log-likelihood gradient vector) = 758.731278 Norm (lambda vector) = 386.928813 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4582 98.12 98.90 98.51 i-np 13660 13602 13356 98.19 97.77 97.98 e-np 12220 12284 12060 98.18 98.69 98.43 o 6349 6382 6202 97.18 97.68 97.43 e-vp 4768 4732 4662 98.52 97.78 98.15 i-vp 2602 2653 2535 95.55 97.43 96.48 e-adjp 384 381 337 88.45 87.76 88.10 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 813 718 88.31 87.35 87.83 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 481 458 95.22 91.05 93.09 i-adjp 152 137 116 84.67 76.32 80.28 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.01 75.50 78.62 Avg2. 46451 46451 45307 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12284 11801 96.07 96.57 96.32 pp 4633 4670 4570 97.86 98.64 98.25 vp 4768 4732 4558 96.32 95.60 95.96 sbar 503 481 451 93.76 89.66 91.67 adjp 384 381 320 83.99 83.33 83.66 advp 822 813 713 87.70 86.74 87.22 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.27 82.90 87.33 Avg2. 23486 23516 22558 95.93 96.05 95.99 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 190 Log-likelihood = -22241.786150 Norm (log-likelihood gradient vector) = 4770.944709 Norm (lambda vector) = 395.088457 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4581 98.12 98.88 98.49 i-np 13660 13683 13403 97.95 98.12 98.04 e-np 12220 12240 12041 98.37 98.54 98.45 o 6349 6348 6187 97.46 97.45 97.46 e-vp 4768 4733 4662 98.50 97.78 98.14 i-vp 2602 2647 2531 95.62 97.27 96.44 e-adjp 384 380 338 88.95 88.02 88.48 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 815 716 87.85 87.10 87.48 i-advp 100 82 71 86.59 71.00 78.02 e-sbar 503 484 458 94.63 91.05 92.81 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.02 75.49 78.62 Avg2. 46451 46451 45314 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12240 11785 96.28 96.44 96.36 pp 4633 4669 4569 97.86 98.62 98.24 vp 4768 4733 4555 96.24 95.53 95.88 sbar 503 484 451 93.18 89.66 91.39 adjp 384 380 320 84.21 83.33 83.77 advp 822 815 711 87.24 86.50 86.87 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.13 82.85 87.24 Avg2. 23486 23477 22536 95.99 95.96 95.97 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 191 Log-likelihood = -21681.281780 Norm (log-likelihood gradient vector) = 1403.592113 Norm (lambda vector) = 389.712384 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4666 4580 98.16 98.86 98.51 i-np 13660 13696 13408 97.90 98.16 98.03 e-np 12220 12236 12038 98.38 98.51 98.45 o 6349 6342 6182 97.48 97.37 97.42 e-vp 4768 4732 4663 98.54 97.80 98.17 i-vp 2602 2647 2532 95.66 97.31 96.48 e-adjp 384 381 338 88.71 88.02 88.37 i-pp 52 44 35 79.55 67.31 72.92 e-advp 822 813 716 88.07 87.10 87.58 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 486 460 94.65 91.45 93.02 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 133 124 93.23 98.41 95.75 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.04 75.51 78.64 Avg2. 46451 46451 45314 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12236 11782 96.29 96.42 96.35 pp 4633 4666 4568 97.90 98.60 98.25 vp 4768 4732 4556 96.28 95.55 95.92 sbar 503 486 453 93.21 90.06 91.61 adjp 384 381 320 83.99 83.33 83.66 advp 822 813 711 87.45 86.50 86.97 prt 126 133 124 93.23 98.41 95.75 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.07 82.89 87.24 Avg2. 23486 23471 22535 96.01 95.95 95.98 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 192 Log-likelihood = -21357.664281 Norm (log-likelihood gradient vector) = 606.544299 Norm (lambda vector) = 393.068400 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4662 4579 98.22 98.83 98.53 i-np 13660 13700 13410 97.88 98.17 98.03 e-np 12220 12238 12039 98.37 98.52 98.45 o 6349 6337 6178 97.49 97.31 97.40 e-vp 4768 4731 4662 98.54 97.78 98.16 i-vp 2602 2651 2534 95.59 97.39 96.48 e-adjp 384 379 337 88.92 87.76 88.34 i-pp 52 45 35 77.78 67.31 72.16 e-advp 822 811 716 88.29 87.10 87.69 i-advp 100 81 71 87.65 71.00 78.45 e-sbar 503 490 463 94.49 92.05 93.25 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 132 124 93.94 98.41 96.12 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.00 75.52 78.63 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12238 11783 96.28 96.42 96.35 pp 4633 4662 4566 97.94 98.55 98.25 vp 4768 4731 4556 96.30 95.55 95.93 sbar 503 490 456 93.06 90.66 91.84 adjp 384 379 320 84.43 83.33 83.88 advp 822 811 711 87.67 86.50 87.08 prt 126 132 124 93.94 98.41 96.12 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.19 82.94 87.32 Avg2. 23486 23467 22537 96.04 95.96 96.00 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 193 Log-likelihood = -21204.230513 Norm (log-likelihood gradient vector) = 490.099053 Norm (lambda vector) = 394.845714 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4663 4580 98.22 98.86 98.54 i-np 13660 13710 13410 97.81 98.17 97.99 e-np 12220 12234 12039 98.41 98.52 98.46 o 6349 6331 6175 97.54 97.26 97.40 e-vp 4768 4731 4661 98.52 97.76 98.14 i-vp 2602 2654 2537 95.59 97.50 96.54 e-adjp 384 379 337 88.92 87.76 88.34 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 808 716 88.61 87.10 87.85 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 490 463 94.49 92.05 93.25 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.03 75.53 78.64 Avg2. 46451 46451 45315 97.55 97.55 97.55 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12234 11781 96.30 96.41 96.35 pp 4633 4663 4567 97.94 98.58 98.26 vp 4768 4731 4556 96.30 95.55 95.93 sbar 503 490 456 93.06 90.66 91.84 adjp 384 379 320 84.43 83.33 83.88 advp 822 808 711 88.00 86.50 87.24 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.30 82.94 87.37 Avg2. 23486 23460 22536 96.06 95.96 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 194 Log-likelihood = -21047.469385 Norm (log-likelihood gradient vector) = 580.237361 Norm (lambda vector) = 396.848743 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 322 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4664 4582 98.24 98.90 98.57 i-np 13660 13711 13417 97.86 98.22 98.04 e-np 12220 12231 12039 98.43 98.52 98.47 o 6349 6335 6178 97.52 97.31 97.41 e-vp 4768 4730 4661 98.54 97.76 98.15 i-vp 2602 2648 2533 95.66 97.35 96.50 e-adjp 384 382 339 88.74 88.28 88.51 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 809 717 88.63 87.23 87.92 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 490 463 94.49 92.05 93.25 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.03 75.56 78.66 Avg2. 46451 46451 45326 97.58 97.58 97.58 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12231 11786 96.36 96.45 96.41 pp 4633 4664 4569 97.96 98.62 98.29 vp 4768 4730 4556 96.32 95.55 95.94 sbar 503 490 456 93.06 90.66 91.84 adjp 384 382 321 84.03 83.59 83.81 advp 822 809 712 88.01 86.62 87.31 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.27 82.99 87.38 Avg2. 23486 23461 22545 96.10 95.99 96.04 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 195 Log-likelihood = -20781.146463 Norm (log-likelihood gradient vector) = 599.814736 Norm (lambda vector) = 400.342112 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4675 4583 98.03 98.92 98.47 i-np 13660 13761 13418 97.51 98.23 97.87 e-np 12220 12212 12018 98.41 98.35 98.38 o 6349 6304 6153 97.60 96.91 97.26 e-vp 4768 4730 4658 98.48 97.69 98.08 i-vp 2602 2646 2530 95.62 97.23 96.42 e-adjp 384 382 339 88.74 88.28 88.51 i-pp 52 49 35 71.43 67.31 69.31 e-advp 822 811 718 88.53 87.35 87.94 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 480 457 95.21 90.85 92.98 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 20 19 95.00 79.17 86.36 e-conjp 16 12 11 91.67 68.75 78.57 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.74 74.70 78.06 Avg2. 46451 46451 45267 97.45 97.45 97.45 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12212 11744 96.17 96.10 96.14 pp 4633 4675 4570 97.75 98.64 98.20 vp 4768 4730 4551 96.22 95.45 95.83 sbar 503 480 450 93.75 89.46 91.56 adjp 384 382 322 84.29 83.85 84.07 advp 822 811 713 87.92 86.74 87.32 prt 126 130 123 94.62 97.62 96.09 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 12 11 91.67 68.75 78.57 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.24 82.16 86.91 Avg2. 23486 23443 22493 95.95 95.77 95.86 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 196 Log-likelihood = -20803.822878 Norm (log-likelihood gradient vector) = 2394.403456 Norm (lambda vector) = 408.077315 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 320 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4669 4582 98.14 98.90 98.52 i-np 13660 13731 13416 97.71 98.21 97.96 e-np 12220 12223 12029 98.41 98.44 98.42 o 6349 6323 6169 97.56 97.16 97.36 e-vp 4768 4730 4660 98.52 97.73 98.13 i-vp 2602 2648 2533 95.66 97.35 96.50 e-adjp 384 382 339 88.74 88.28 88.51 i-pp 52 47 35 74.47 67.31 70.71 e-advp 822 809 718 88.75 87.35 88.04 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 484 459 94.83 91.25 93.01 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 81.96 75.51 78.60 Avg2. 46451 46451 45302 97.53 97.53 97.53 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12223 11764 96.24 96.27 96.26 pp 4633 4669 4569 97.86 98.62 98.24 vp 4768 4730 4555 96.30 95.53 95.91 sbar 503 484 452 93.39 89.86 91.59 adjp 384 382 322 84.29 83.85 84.07 advp 822 809 713 88.13 86.74 87.43 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.32 82.93 87.37 Avg2. 23486 23452 22520 96.03 95.89 95.96 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 197 Log-likelihood = -20625.159259 Norm (log-likelihood gradient vector) = 1148.364020 Norm (lambda vector) = 404.150345 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4581 98.16 98.88 98.52 i-np 13660 13708 13416 97.87 98.21 98.04 e-np 12220 12230 12038 98.43 98.51 98.47 o 6349 6335 6181 97.57 97.35 97.46 e-vp 4768 4730 4659 98.50 97.71 98.10 i-vp 2602 2645 2530 95.65 97.23 96.44 e-adjp 384 385 340 88.31 88.54 88.43 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 814 719 88.33 87.47 87.90 i-advp 100 80 71 88.75 71.00 78.89 e-sbar 503 487 462 94.87 91.85 93.33 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.00 75.53 78.63 Avg2. 46451 46451 45322 97.57 97.57 97.57 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12230 11784 96.35 96.43 96.39 pp 4633 4667 4569 97.90 98.62 98.26 vp 4768 4730 4552 96.24 95.47 95.85 sbar 503 487 455 93.43 90.46 91.92 adjp 384 385 322 83.64 83.85 83.75 advp 822 814 714 87.71 86.86 87.29 prt 126 130 123 94.62 97.62 96.09 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.22 82.93 87.33 Avg2. 23486 23467 22540 96.05 95.97 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 198 Log-likelihood = -20389.629749 Norm (log-likelihood gradient vector) = 548.465383 Norm (lambda vector) = 406.858672 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4668 4580 98.11 98.86 98.48 i-np 13660 13717 13418 97.82 98.23 98.02 e-np 12220 12225 12035 98.45 98.49 98.47 o 6349 6335 6182 97.58 97.37 97.48 e-vp 4768 4732 4659 98.46 97.71 98.08 i-vp 2602 2645 2528 95.58 97.16 96.36 e-adjp 384 383 339 88.51 88.28 88.40 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 812 718 88.42 87.35 87.88 i-advp 100 79 71 89.87 71.00 79.33 e-sbar 503 485 461 95.05 91.65 93.32 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.07 75.49 78.65 Avg2. 46451 46451 45316 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12225 11781 96.37 96.41 96.39 pp 4633 4668 4568 97.86 98.60 98.23 vp 4768 4732 4551 96.17 95.45 95.81 sbar 503 485 454 93.61 90.26 91.90 adjp 384 383 321 83.81 83.59 83.70 advp 822 812 713 87.81 86.74 87.27 prt 126 130 123 94.62 97.62 96.09 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.26 82.87 87.31 Avg2. 23486 23459 22532 96.05 95.94 95.99 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds Iteration: 199 Log-likelihood = -20223.894199 Norm (log-likelihood gradient vector) = 431.458828 Norm (lambda vector) = 408.615220 Log-likelihood and gradient computational time: 320 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4670 4580 98.07 98.86 98.46 i-np 13660 13646 13382 98.07 97.96 98.02 e-np 12220 12261 12051 98.29 98.62 98.45 o 6349 6370 6195 97.25 97.57 97.41 e-vp 4768 4733 4659 98.44 97.71 98.07 i-vp 2602 2646 2529 95.58 97.19 96.38 e-adjp 384 383 338 88.25 88.02 88.14 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 809 717 88.63 87.23 87.92 i-advp 100 78 70 89.74 70.00 78.65 e-sbar 503 484 460 95.04 91.45 93.21 i-adjp 152 136 116 85.29 76.32 80.56 e-prt 126 130 123 94.62 97.62 96.09 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.02 75.42 78.58 Avg2. 46451 46451 45306 97.54 97.54 97.54 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12261 11799 96.23 96.55 96.39 pp 4633 4670 4568 97.82 98.60 98.20 vp 4768 4733 4552 96.18 95.47 95.82 sbar 503 484 453 93.60 90.06 91.79 adjp 384 383 320 83.55 83.33 83.44 advp 822 809 712 88.01 86.62 87.31 prt 126 130 123 94.62 97.62 96.09 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.23 82.83 87.28 Avg2. 23486 23494 22548 95.97 96.01 95.99 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 355 seconds Iteration: 200 Log-likelihood = -20081.814179 Norm (log-likelihood gradient vector) = 1283.853495 Norm (lambda vector) = 411.698842 Log-likelihood and gradient computational time: 321 seconds Training iteration elapsed: 321 seconds Label-based performance evaluation: Label Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- e-pp 4633 4667 4579 98.11 98.83 98.47 i-np 13660 13701 13411 97.88 98.18 98.03 e-np 12220 12237 12042 98.41 98.54 98.47 o 6349 6342 6186 97.54 97.43 97.49 e-vp 4768 4733 4659 98.44 97.71 98.07 i-vp 2602 2647 2530 95.58 97.23 96.40 e-adjp 384 382 338 88.48 88.02 88.25 i-pp 52 46 35 76.09 67.31 71.43 e-advp 822 807 716 88.72 87.10 87.91 i-advp 100 78 70 89.74 70.00 78.65 e-sbar 503 486 461 94.86 91.65 93.23 i-adjp 152 135 116 85.93 76.32 80.84 e-prt 126 131 124 94.66 98.41 96.50 i-sbar 12 13 9 69.23 75.00 72.00 i-conjp 24 22 21 95.45 87.50 91.30 e-conjp 16 13 12 92.31 75.00 82.76 e-intj 4 1 1 100.00 25.00 40.00 e-lst 10 10 8 80.00 80.00 80.00 i-ucp 0 0 0 0.00 0.00 0.00 e-ucp 0 0 0 0.00 0.00 0.00 i-intj 12 0 0 0.00 0.00 0.00 i-prt 0 0 0 0.00 0.00 0.00 i-lst 2 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 82.07 75.46 78.63 Avg2. 46451 46451 45318 97.56 97.56 97.56 Chunk-based performance evaluation: Chunk Manual Model Match Pre.(%) Rec.(%) F1-Measure(%) ----- ------ ----- ----- ------- ------- ------------- np 12220 12237 11790 96.35 96.48 96.41 pp 4633 4667 4567 97.86 98.58 98.22 vp 4768 4733 4553 96.20 95.49 95.84 sbar 503 486 454 93.42 90.26 91.81 adjp 384 382 320 83.77 83.33 83.55 advp 822 807 711 88.10 86.50 87.29 prt 126 131 124 94.66 98.41 96.50 lst 10 10 8 80.00 80.00 80.00 intj 4 1 1 100.00 25.00 40.00 conjp 16 13 12 92.31 75.00 82.76 ucp 0 0 0 0.00 0.00 0.00 ----- ------ ----- ----- ------- ------- ------------- Avg1. 92.27 82.90 87.34 Avg2. 23486 23467 22540 96.05 95.97 96.01 Current max chunk-based F1: 96.18 (iteration 110) Training iteration elapsed (including evaluation time): 356 seconds The training process elapsed: 71111 seconds