word segmentation 0.003494839
chinese word 0.003400325
joint word 0.003284216
work word 0.003139523
word seg 0.003036742
multiple word 0.003011253
supervised word 0.002955578
whole word 0.002911602
pos tag 0.00289707
word segmen 0.002887876
pos tagging 0.002698596
pos information 0.0025141539999999998
character feature 0.00251221
pos tags 0.002508408
training data 0.00220448
window features 0.0021588140000000002
additional features 0.002126614
linear model 0.0020989520000000003
clude pos 0.0020747689999999997
linguistic features 0.00206373
features lookup 0.002061774
chinese character 0.002061045
useful features 0.0019988740000000003
relevant features 0.00198489
character tags 0.001973198
extracts features 0.001967579
gazetteer features 0.001962049
different character 0.001956849
linear tag 0.001942482
training algorithm 0.001920667
training corpus 0.001914402
joint model 0.001907966
chinese words 0.0019061249999999998
training method 0.001894863
feature vector 0.001799995
character position 0.001762964
language model 0.001738781
training set 0.001725089
training problem 0.001710363
features 0.00169909
new training 0.001691743
tag sequence 0.001682472
additional feature 0.0016620839999999999
training time 0.001661823
possible tag 0.0016569279999999998
character dictionary 0.001654936
character embeddings 0.001651826
character lookup 0.001640334
feature values 0.00163309
feature vectors 0.001631114
feature templates 0.001630995
training example 0.001625954
tagging approach 0.0016196700000000001
character representations 0.001609272
single character 0.0016013169999999999
training process 0.001599866
supervised training 0.001587598
trained model 0.001587578
useful character 0.001577434
feature engineering 0.001560356
tag set 0.001560349
tag inference 0.001546256
unsupervised training 0.0015428949999999999
training sets 0.001542723
feature dimension 0.001542034
training stage 0.0015402459999999999
central character 0.001540125
tain character 0.001540125
chinese characters 0.0015398220000000002
training pairs 0.001537638
ear model 0.00153184
training algo 0.001519343
learning algorithm 0.001514235
crfs model 0.001513103
training criterion 0.001511342
training setr 0.001511342
guage model 0.001503844
feature extraction 0.001500444
corresponding tag 0.001490487
test data 0.001488583
tagging models 0.0014846310000000001
sequence tagging 0.001483998
tagging tasks 0.001451627
unknown words 0.001450307
data set 0.001431669
tag path 0.0014039859999999999
correct tag 0.001385008
different tags 0.001374747
same data 0.001371977
tag paths 0.001364571
ditional tag 0.0013517759999999998
tagging problem 0.001347149
perceptron algorithm 0.001330987
unlabeled data 0.001325397
segmentation systems 0.001294782
character 0.00127765
linear time 0.001271145
other tags 0.0012654979999999999
tagging task 0.001260966
new function 0.00125292
