manning schuetze statisticalnlp phần 5 pdf

... Acquisition P(c) P(cleat) P(clsee1 P(clfind) 0. 25 0.01 0. 25 0.33 0. 25 0.01 0. 25 0.33 0. 25 0.97 0. 25 0.33 0. 25 0.01 0. 25 0.01 1.76 0.00 0. 35 Table 8 .5 Selectional Preference Strength (SPS). The ... data in figure 8 .5. Pl = P(SovietIcosmonaut) = 0 .5 P2 = 0 173 = P (spacewalking cosmonaut) = 0 .5 41 = 0 92 = P(AmericanIastronuut) = 0 .5 93 = P(spacewulkingIa...

Ngày tải lên: 14/08/2014, 08:22

... results for an 1NN categorizer for the “earnings” category. xix 54 2 54 2 54 4 54 7 55 0 55 4 56 2 57 6 57 7 58 1 58 2 58 6 59 3 59 4 59 5 59 5 601 602 606 Preface . xxxlll in an accessible fashion will create ... models 55 4 15. 4 Latent Semantic Indexing 55 4 15. 4.1 Least-squares methods 55 7 15. 4.2 Singular Value Decomposition 55 8 15. 4.3 Latent Semantic Inde...

Ngày tải lên: 14/08/2014, 08:22

... tags Brown 87 179 Penn 45 CLAWS1 132 CLAWS2 166 CLAWS c5 62 London-Lund 197 Table 4.4 Sizes of various tag sets. C5TAGSET Corpus (CLAWS1 through CLAWS5; CLAWS5 is also referred to as the c5 PENNTREEBANKTAG tag ... experimentally tested set of tokenization rules is the set used for the Susanne corpus (Sampson 19 95: 52 -59 ). Nunberg (1990) provides a linguistic perspective on the import...

Ngày tải lên: 14/08/2014, 08:22

... parse Precision: = 37 .5% Crossing Brackets: 0 Recall: = 37 .5% Crossing Accuracy: 100% Labeled Precision: = 37 .5% Tagging Accuracy: 1 = 90.9% Labeled Recall: = 37 .5% Figure 12.6An example ... 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...

Ngày tải lên: 21/08/2013, 10:59

... lo7 2 8 244.42 4 16 1.34 x 10s 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 ... -H(P) Therefore, nl$~r(x; Y) = 1 -H(p) I 1 f r .r i- :. 1, Y, tY J: (2.47) (2.48) (2.49) (2 .50 ) (2 .51 ) 2.2 Essential Information Theory 7 .5 = Ep (log &...

Ngày tải lên: 14/08/2014, 08:22

... 1 System 2 scores 71, 61, 55 , 60, 68, 49, 42, 55 , 75, 45, 54 , 51 42, 72, 76, 55 , 64 55 , 36, 58 , 55 , 67 total 609 52 6 n 11 11 mean Xi 55 .4 47.8 Sf = C(Xi,j - Xi)2 1,3 75. 4 1,228.g df 10 10 Pooled ... Data Bigrams r N, r 1 138741 28 2 254 13 29 3 1 053 1 30 4 59 97 31 5 356 5 32 6 2486 7 1 754 1264 8 1342 1366 9 1106 1917 10 896 2233 . . . 250 7 NV 90 120 86 9...

Ngày tải lên: 14/08/2014, 08:22

... 3 75 Correct Tags assigned by the tagger Tags DT IN JJ NN RB RP VB VBG DT 99.4 .3 .3 IN .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 ... PERIOD 0 0 0 48636 0 19 1973 0 426 187 0 38 43322 0 13 25 17314 0 1 85 1067 3720 42470 11773 614 21392 6072 42 4 758 1476 129 152 2 8016 75 4 656 1329 954 0 First tag AT BEZ IN...

Ngày tải lên: 14/08/2014, 08:22

... (1998), and Wu and Wong 14.1 Hierarchical Clustering 50 5 0 1 2 3 4 5 6 7 8 Figure 14.6 Single-link clustering of the points in figure 14.4. 0 1 2 3 4 5 6 7 8 Figure 14.7 Complete-link clustering of ... - 1) 50 4 14 Clustering A 5 4 3 2 1 ax bX -d 4 2d ex fX $d d X tlx 0 ’ I I I I I 1 I I I I I I I I I I * 0 1 2 3 4 5 6 7 8 Figure 14.4 A cloud of points in a plane. 0...

Ngày tải lên: 14/08/2014, 08:22

... surrounding gaps. If cohesion 55 0 15 Topics in Information Retrieval Word k 0 1 2 3 4 5 6 7 8 r9 follows act. 57 552 .0 20142.0 14 35. 0 148.0 18.0 1.0 est. 57 552 .0 20091.0 152 7.3 116.1 8.8 0.7 0.1 0.0 0.0 ... 299.7 230.1 students act. 74343.0 252 3.0 761.0 413.0 2 65. 0 178.0 143.0’ 112.0 96.0 462.0 est. 74343.0 154 0 .5 1061.4 731.3 50 3.8 347.1 239.2 164.8 113 .5 78.2 james a...

Ngày tải lên: 14/08/2014, 08:22

... (1991), 376,414, 451 Abney (1996a), 3 75, 376 Abney (1996b), 34 Abney (19971, 457 absolute discounting, 2 16 accuracy, 269, 57 7 in tagging, 371 accusative, 85 Ackley et al. (19 85) , 402 active ... Dumais, and Gavin W. O ’B rien. 19 95. Using linear algebra for intelligent information retrieval. SIAMReview 37 :57 3 -59 5. Berry, Michael W., and Paul G. Young. 19 95. Using lat...

Ngày tải lên: 14/08/2014, 08:22