word features 0.00471471
word segmentation 0.00423253
chinese word 0.004149896
method word 0.0039476
current word 0.0036640369999999998
entropy word 0.003653588
word seg 0.003624411
new word 0.003617843
word boundary 0.003615616
word context 0.003600545
word candidate 0.003585749
word segmenter 0.0035828649999999997
last word 0.003574047
only word 0.003563887
word boundaries 0.0035633039999999998
average word 0.003536559
word basis 0.003535977
segmented word 0.003521063
unknown word 0.00351566
single word 0.003502935
word segmenters 0.0034878269999999998
word list 0.003472722
official word 0.0034679539999999997
improved word 0.003459354
pos tag 0.00324424
character features 0.00311277
chinese pos 0.0030188660000000003
different pos 0.002844132
pos tags 0.0028235359999999998
pos tagging 0.002804297
tag features 0.0027654000000000003
pos accuracy 0.002706625
same pos 0.002688137
pos tagger 0.0026439330000000002
pos ambiguity 0.002583635
chinese character 0.002547956
good pos 0.002476934
english pos 0.002432387
average pos 0.002405529
correct pos 0.002393338
tag feature 0.00237484
subsequent pos 0.00235495
probable pos 0.002334797
pos taggers 0.002330909
unique pos 0.002329136
detailed pos 0.002327974
character sequence 0.00230567
chinese words 0.002269456
first character 0.002208991
previous character 0.002106581
lexical features 0.002104756
input character 0.002091127
additional features 0.0020674200000000004
current character 0.002062097
character position 0.002001635
segmenter features 0.001972995
last character 0.001972107
tag sequence 0.0019583
individual character 0.001933512
tag information 0.0019151279999999999
single character 0.001900995
training corpus 0.001900616
training data 0.0018160810000000002
tag set 0.001761135
segmentation accuracy 0.0017456050000000001
chinese characters 0.001668915
boundary tag 0.001666306
feature templates 0.001641031
segmented words 0.001640623
feature representation 0.001638773
surrounding words 0.0016166379999999999
training set 0.001611535
tag prediction 0.001604694
consecutive words 0.001592108
treebank tag 0.001592038
additional training 0.00157838
markov model 0.001571267
ctb training 0.001562208
character 0.00156035
tagging method 0.0015583469999999999
following feature 0.001553011
features 0.00155242
tag assignment 0.001545113
chinese sentence 0.001537946
chinese parser 0.001528552
training time 0.001527213
tag sequences 0.001525967
feature cutoff 0.001512154
valid tag 0.001512124
assigned tag 0.001510674
chinese language 0.00147293
default feature 0.0014596890000000001
training text 0.001456785
tagging accuracy 0.0014484020000000001
average training 0.001437649
sufficient training 0.0014084170000000001
segmentation bakeoff 0.001395042
penn chinese 0.001383302
segmentation errors 0.001382136
segmentation scorer 0.00136712
