manning schuetze statisticalnlp phần 1 pdf

... Areas 11 2 3.5 Further Reading 11 3 3.6 Exercises 11 4 4 Corpus-Based Work 11 7 4 .1 Getting Set Up 11 8 4 .1. 1 Computers 11 8 4 .1. 2 Corpora 11 8 4 .1. 3 Software 12 0 4.2 Looking at Text 12 3 4.2 .1 Low-level ... sentence 396 11 .3.4 Training a PCFG 398 11 .4 Problems with the Inside-Outside Algorithm 4 01 11. 5 Further Reading 402 11 .6 Exercises 404 12 Probabilistic Par...

Ngày tải lên: 14/08/2014, 08:22

... Treebank PRP VBD VBN IN DT NN MD VB PRP ICE PRON(pers,sing) AUX(pass,past) V(ditr,edp) CONJUNC(subord) ART(def) N(com,sing) AUX(modal,past) V(montr,infin) PRON(poss,sing) PUNC(per1 Figure 4.2 A sentence as tagged according to several different tag sets. Tag set Basic size Total tags Brown 87 17 9 Penn 45 CLAWS1 13 2 CLAWS2 16 6 CLAWS c5 62 London-Lund 19 7 Table ... Amer- ican English f...

Ngày tải lên: 14/08/2014, 08:22

... their meaning. (See (Apresjan 19 74), (Pustejovsky 19 91) , (Lakoff 19 87), (Ostler and Atkins 19 9 21, (Nunberg and Zaenen 19 92), and (Copes- take and Briscoe 19 95) for theoretical work on ... (Pereira et al. 19 93; Zernik 19 9lb; Dolan 19 94; Pedersen and Bruce 19 97; Chen and Chang 19 98). Pereira et al, (19 93) cluster contexts of words in a way similar to Schutze...

Ngày tải lên: 14/08/2014, 08:22

... DT NN NP NNP NP NN NP JJ NN NP NP SBAR 1st Obj 7.5% 13 .4% 12 .2% 10 .4% 4.5% 3.9% 1. 1% 0.3% % as 2nd Obj 0.2% 0.9% 14 .4% 13 .3% 5.9% 9.2% 10 .4% 5 .1% Table 12 .4 Selected common expansions of NP as ... the signal. 12 .1 Some Concepts 427 (12 .17 ) = attach) = gc) if top is not gc 0 otherwise (12 .18 ) = = if top is not gc 0 otherwise Where these operations obey the foll...

Ngày tải lên: 21/08/2013, 10:59

... lo7 2 8 244.42 4 16 1. 34 x 10 s 3 7 36. 21 6 14 2307.06 4 6 7.54 8 12 87.89 5 5 2 .16 10 10 6.89 6 4 0.84 12 8 1. 09 7 3 0.45 14 6 0.35 8 2 0.36 16 4 0.25 9 1 0.37 18 2 0.48 10 0 0.68 20 0 3.74 Table 2 .1 Likelihood ratios ... efficient way is to simply encode the result as a 3 digit binary message: 1 2 3 4 5 6 7 8 0 01 010 011 10 0 10 1 11 0 11 1 000 The tran...

Ngày tải lên: 14/08/2014, 08:22

... . 2507 NV 90 12 0 86 98 99 . . . 1 1 1 1 1 Y 1 2 3 4 5 6 7 8 9 10 Trigrams N, Y 404 211 28 32 514 29 10 056 30 4780 31 24 91 32 15 71 1088 18 9 749 202 582 214 432 366 . . . 378 Nr 35 32 25 18 19 . . 1 1 1 1 1 Table ... meaningful 214 6 Statistical Inference: n-gram Models over Sparse Data Bigrams r N, r 1 1387 41 28 2 25 413 29 3 10 5 31 30 4...

Ngày tải lên: 14/08/2014, 08:22

... PERIOD 0 0 0 48636 0 19 19 73 0 426 18 7 0 38 43322 0 13 25 17 314 0 18 5 10 67 3720 42470 11 773 614 213 92 6072 42 4758 14 76 12 9 15 22 8 016 75 4656 13 29 954 0 First tag AT BEZ IN NN VB PERIOD Table 10 .3 Idealized ... .4 97.5 1. 5 .5 JJ .l 93.9 1. 8 .9 .l .4 NN 2.2 95.5 .2 .4 RB .2 2.4 2.2 .6 93.2 1. 2 RP 24.7 1. 1 12 .6 61. 5 VB .3 1. 4 96.0 VBG 2.5 4.4 93.0 Tabl...

Ngày tải lên: 14/08/2014, 08:22

... word:2 (14 .10 ) H(L,7T) ” & 1 C(w1w2) logP(c2lcdPwlc2) w1w* Formula (14 .10 ) can be simplified as follows: (14 .11 ) H(L,?T) E - c [ c(w1w2) [logP(w2(c2) + logP(c2)] w1w2 N -1 2. ... al. (19 98). Alshawi et al. (19 97), Wang and Waibel (19 98), and Wu and Wong 14 .1 Hierarchical Clustering 505 0 1 2 3 4 5 6 7 8 Figure 14 .6 Single-link clustering o...

Ngày tải lên: 14/08/2014, 08:22

... 7 61. 0 413 .0 265.0 17 8.0 14 3.0’ 11 2.0 96.0 462.0 est. 74343.0 15 40.5 10 61. 4 7 31. 3 503.8 347 .1 239.2 16 4.8 11 3.5 78.2 james act. 7 010 5.0 7953.0 922.0 18 3.0 52.0 24.0 19 .0 9.0 7.0 22.0 est. 7 010 5.0 ... If cohesion 550 15 Topics in Information Retrieval Word k 0 1 2 3 4 5 6 7 8 r9 follows act. 57552.0 2 014 2.0 14 35.0 14 8.0 18 .0 1. 0 est. 57552.0 200 91. 0...

Ngày tải lên: 14/08/2014, 08:22

... 384 Anderson (19 8 31, 35 Anderson (19 9 01, 3 5 annealing schedule, 479 antonyms, 11 0 Aone and McKee (19 95), 313 Appelt et al. (19 9 31, 314 Apresjan (19 74), 258 Apte et al. (19 94), 579 arc-emission ... Linguistics 19 :3 31- 358. Qiu, Yonggang, and H.P. Frei. 19 93. Concept based query expansion. In SZGZR ‘9 3, pp. 16 0 -16 9. Quinlan, J. R. 19 86. Induction...

Ngày tải lên: 14/08/2014, 08:22