model segmentation 0.003097962
joint model 0.002842267
supervised model 0.002777964
current model 0.002690505
crfs model 0.002674979
model parameters 0.002641568
sequence model 0.002629469
cws model 0.002607232
sequential model 0.002567923
word segmentation 0.002564092
model induction 0.002536701
our model 0.002516833
pipeline model 0.0025036290000000003
general model 0.002491549
unlabeled data 0.002298003
chinese word 0.0022509599999999998
test data 0.002246608
model 0.00222543
pos tag 0.0021804
word boundary 0.002088252
right word 0.002085216
labeled data 0.002072511
data size 0.002040525
center word 0.00203879
ing data 0.001975396
feature function 0.001902567
same feature 0.0018960079999999998
segmentation pos 0.001895632
data sources 0.0018794060000000001
train data 0.001872301
experimental data 0.001833088
testing data 0.0018199560000000002
pos tags 0.0017863240000000002
data sizes 0.001778627
velopment data 0.0017734740000000001
baseline feature 0.001679498
pos tagging 0.001647814
feature weights 0.0016155929999999998
feature trigram 0.0016057579999999999
description feature 0.001492544
evidence feature 0.0014908599999999999
feature templates 0.001479998
other words 0.0014783399999999999
novel feature 0.0014714619999999998
supervised learning 0.001459805
feature instantiation 0.001455987
feature tem 0.001445779
same label 0.001436819
pos sequence 0.001427139
joint models 0.0014145450000000001
similar words 0.0014098539999999999
graph information 0.001406455
true tag 0.001402285
rect tag 0.001402285
chinese language 0.0014012130000000001
learning objective 0.001379794
label information 0.001370739
pos part 0.0013327620000000001
segmentation error 0.001318728
new features 0.001299021
unlabeled size 0.001280708
similarity graph 0.001268587
quent pos 0.001267542
learning problem 0.001261535
segmented words 0.001260387
similar label 0.0012534199999999999
crfs models 0.001247257
label distribution 0.001228997
learning curve 0.001218217
test set 0.001209247
crfs training 0.001204935
label distributions 0.001203467
similarity function 0.001199227
feature 0.0011977
natural language 0.001191298
statistical features 0.00118634
graph propagation 0.0011856190000000002
lexicon features 0.001181266
objective function 0.00117739
unlabeled sentences 0.001159051
unstructured learning 0.0011524159999999999
label propagation 0.0011499029999999999
oov words 0.00114855
rare features 0.001147931
correlated features 0.001143919
label sequence 0.00114255
lion words 0.0011414189999999999
language processing 0.001138461
label constraints 0.001130513
segmentation ambiguities 0.0011263129999999999
graph con 0.001121039
same parameters 0.001114446
weighted graph 0.001104275
syntactic information 0.001098204
specific label 0.0010853619999999999
unlabeled examples 0.001074237
graphical models 0.001073795
prior knowledge 0.001072999
entire graph 0.0010721110000000002
tagging error 0.00107091
