word segmentation 0.003980488
chinese word 0.00381894
word tree 0.003618614
word parse 0.00353955
current word 0.003463412
baseline word 0.003462966
corresponding word 0.003422901
word dictionary 0.003388715
only word 0.003387493
word boundary 0.0033848560000000003
word level 0.0033756280000000003
word seg 0.00336426
word length 0.00335062
word boundaries 0.003308035
word list 0.003293275
word segmen 0.00328699
word rate 0.003274345
word classes 0.003273262
word segmentations 0.003268922
ambiguous word 0.003260887
word dic 0.003260344
tag model 0.002508532
tag feature 0.002330612
pos tag 0.002220952
chinese words 0.0022191
language model 0.002128951
pos information 0.00202586
baseline model 0.002021776
other words 0.002015
model knowledge 0.0019967929999999997
chinese character 0.00196825
pos tags 0.0019372489999999998
entropy model 0.001920916
component model 0.0019115339999999999
pos tagging 0.0019100649999999999
lexical features 0.001895082
feature function 0.0018947960000000002
feature templates 0.0018550230000000001
check model 0.001820408
character tree 0.001767924
common words 0.0017586339999999998
training data 0.001757341
compound words 0.001749598
character sentence 0.001727574
unknown words 0.001719081
independent feature 0.001695715
same training 0.0016895579999999999
independent features 0.0016835449999999998
unseen words 0.00168256
training corpus 0.001674614
first character 0.001673623
functional words 0.0016621409999999998
chunk feature 0.0016619310000000002
boundary pos 0.001656086
level pos 0.001646858
check feature 0.001642488
feature func 0.001641876
segmentation error 0.001637844
ical features 0.001630983
features extend 0.001630983
extend features 0.001630983
features most 0.0016306369999999999
present features 0.001629179
segmentation label 0.001619006
character parser 0.001590395
model 0.00156129
pos info 0.001533632
character level 0.001524938
last character 0.001524564
character trees 0.001489781
training set 0.00147959
nese character 0.001471733
training sentences 0.001442855
character vocabulary 0.001434261
current training 0.001430364
ning character 0.001412851
character streams 0.001410904
character emf 0.001409435
character tokenization 0.001409435
words 0.00140264
training time 0.001402
chinese characters 0.0013944040000000001
label information 0.001393148
chinese language 0.0013841209999999999
feature 0.00138337
features 0.0013712
training size 0.0013581700000000001
statistical tag 0.001345222
segmentation experiment 0.001338998
parsing accuracy 0.001325985
experiments segmentation 0.001322823
multiple tag 0.00132069
syntactic information 0.0013048740000000001
segmented training 0.001301957
segmentation algorithms 0.001301847
parsing label 0.001293626
training algorithms 0.001293271
efficient training 0.001263162
few segmentation 0.001260659
initial segmentation 0.001242009
