domain words 0.0030831590000000002
such words 0.0030003020000000003
english words 0.002866223
domain word 0.002858589
words context 0.002831193
such word 0.002775732
new words 0.002774083
ing words 0.002658082
similar words 0.002632548
wsj words 0.002623627
other word 0.00262014
related words 0.0025318190000000003
unknown words 0.00250014
frequency words 0.0024822570000000003
corresponding words 0.002462381
frequent words 0.002457827
prefix words 0.002457223
coded words 0.002441637
nearby words 0.0024213330000000003
quent words 0.002420152
hyphenated words 0.002420152
word contexts 0.0023159870000000003
word type 0.002301751
pos tag 0.0022974
stem word 0.002286783
frequent word 0.002233257
words 0.00216407
pos tags 0.002080646
initial model 0.002028293
tag information 0.001970469
learning model 0.001775435
domain text 0.0017411430000000001
pos tagging 0.001735748
tag learning 0.001729735
pos tagger 0.0017015939999999998
tagging model 0.001700408
such information 0.0016985210000000001
tagger model 0.0016662539999999998
possible pos 0.001657974
similar pos 0.001657698
hmm model 0.001657141
tag probability 0.0016547150000000002
text corpus 0.001649232
supervised pos 0.001610938
good tag 0.00156386
correct tag 0.00156083
current model 0.001552178
tag accuracy 0.001529579
new domain 0.001529102
pos category 0.001520244
same domain 0.0015114080000000001
same tags 0.0014837449999999999
large text 0.0014806189999999999
markov model 0.0014774179999999999
model complexity 0.001470456
training data 0.001467776
pos preferences 0.0014664539999999998
verbal pos 0.0014558079999999998
pos taggers 0.001452152
trained model 0.001443911
tag distributions 0.001436457
domain dictionary 0.001434963
model development 0.001431865
actual tag 0.001429158
initial probabilities 0.001421564
final model 0.00141694
domain lexicon 0.00140046
tag informa 0.0013958260000000002
tag sequences 0.0013894570000000002
unigram tag 0.001389354
able tag 0.0013844600000000001
suffix information 0.001375466
probable tag 0.001374818
tags lexicon 0.001372797
surface features 0.001364765
sym tag 0.0013645740000000001
xerox tag 0.0013645740000000001
possible tags 0.00136018
initial lexicon 0.001355784
ing data 0.001336331
such suffixes 0.001297994
specific domain 0.001289181
lexical probabilities 0.0012891489999999999
wsj corpus 0.001286735
wsj text 0.001281611
evaluation domain 0.0012700020000000001
biology domain 0.001262981
medical domain 0.001241729
training set 0.001224386
much information 0.001223255
english dictionary 0.001218027
same training 0.001217776
ogy domain 0.001216963
test set 0.001215632
annotated information 0.001206154
contextual information 0.001202462
domain adaptation 0.0012014
vector information 0.0012003320000000001
sufficient information 0.001199337
initial bigram 0.001196719
