word segmentation 0.00378372
different word 0.00371044
chinese word 0.003613068
word sequence 0.0033455480000000003
ing word 0.003098716
word segmenter 0.0030745160000000002
word boundary 0.0030412110000000003
word clusters 0.0030043910000000004
end word 0.002975719
ferent word 0.002954291
word segmen 0.002949002
start word 0.00294035
automatic word 0.002936211
incompatible word 0.002932123
word segmenta 0.0029253210000000003
style word 0.0029223920000000002
word cek 0.002920377
word spanning 0.0029188160000000003
word segmention 0.0029188160000000003
word delimiters 0.0029188160000000003
word cluster 0.0029188160000000003
different pos 0.00241991
pos tag 0.0023624
chinese words 0.002291328
training corpus 0.002204511
different segmentation 0.0021958000000000004
pos information 0.002164401
pos tags 0.002077671
pos tagging 0.0020444440000000003
model mapping 0.002039048
standard words 0.001932352
pos tagger 0.001926194
other model 0.00191618
character sequence 0.001902418
perceptron training 0.001900759
pos set 0.0018941550000000002
training data 0.00187513
segmented words 0.001764
joint segmentation 0.0017639770000000001
training set 0.0017568649999999998
segmentation standard 0.001739452
baseline model 0.001733042
annotation adaptation 0.001728356
new features 0.0017234659999999999
pos sets 0.001720577
annotation standard 0.0017163319999999999
training algorithm 0.001707403
baseline character 0.001705802
tag sequence 0.001700118
online training 0.0016629149999999998
same character 0.001660164
character classification 0.0016506379999999998
input character 0.001644886
pos tagsets 0.0016435080000000001
baseline features 0.001639162
baseline segmentation 0.001634292
pos postfix 0.001628113
second model 0.001625746
current character 0.001620658
additional training 0.0016043249999999998
discriminative model 0.001601553
training procedure 0.0016001849999999998
training example 0.001598677
different baseline 0.001561012
ing annotation 0.0015609559999999998
feature vector 0.001559912
annotation classification 0.0015560079999999998
different annotations 0.001548735
features source 0.001546109
basic features 0.001544236
chinese parsing 0.001542907
end character 0.001532589
features target 0.001525385
additional features 0.0015223749999999999
third character 0.001519463
source annotation 0.0015181189999999998
only segmentation 0.00150953
guide features 0.001498307
target annotation 0.0014973949999999999
start character 0.00149722
chinese sentence 0.0014878209999999998
segmentation result 0.001487568
ith character 0.0014824179999999999
special features 0.0014819429999999999
character clas 0.001480203
character shares 0.0014762669999999999
specific features 0.001474443
perceptron learning 0.0014600540000000001
normal features 0.001459806
complicated features 0.001447705
feature templates 0.001444345
segmentation standards 0.0014427390000000002
same corpus 0.001437265
iterations segmentation 0.00143278
cial features 0.00142991
annotation standards 0.001419619
features now 0.001410706
tag representation 0.001408248
learning method 0.001402049
automatic annotation 0.0013984509999999998
