label features 0.003266839
function features 0.003127312
feature set 0.003099059
language model 0.003058209
such feature 0.003038715
feature vector 0.00302347
such features 0.003015485
crf model 0.003012924
first feature 0.002961806
feature value 0.002958995
feature values 0.002909887
generative feature 0.002908329
sequence model 0.002906713
markov features 0.002900901
generative features 0.002885099
different model 0.0028841450000000003
feature type 0.002881212
local feature 0.002872633
feature templates 0.002869346
additional features 0.002865745
type features 0.0028579820000000002
local features 0.002849403
main features 0.002828151
feature sets 0.002817628
bigram features 0.002815086
feature map 0.002813963
global feature 0.002796478
above features 0.002794501
certain features 0.00279359
feature computation 0.002768685
corresponding feature 0.002765896
indicator features 0.002765044
odds feature 0.002762967
powerful feature 0.002762084
feature vec 0.002751009
ous feature 0.0027508009999999998
feature mapping 0.002743049
segmentation model 0.002740341
feature decomposition 0.002739654
identity features 0.002719541
valuable features 0.002717312
troublesome features 0.002716239
erful features 0.002716239
generative model 0.002704579
discriminative model 0.002644458
hybrid model 0.002609756
optimal model 0.002605762
baseline model 0.002593439
equivalent model 0.002568282
regression model 0.002565107
model con 0.00255139
tive model 0.002547663
superior model 0.002537695
model parametersw 0.002535716
feature 0.00250812
features 0.00248489
model 0.00230437
training corpus 0.001974084
word segmentation 0.001884201
training data 0.001852673
maximum word 0.001842683
single word 0.00183154
chinese word 0.001822136
word length 0.001790907
particular word 0.00176825
certain word 0.00175693
word tokens 0.001717859
word boundaries 0.001703315
word seg 0.001700432
word identity 0.001682881
word reduplication 0.001679263
training set 0.0016301689999999999
time training 0.0015908569999999998
new words 0.001510876
chinese words 0.001481616
other models 0.001457192
sequence data 0.001415786
training likelihood 0.0013944309999999998
label sequence 0.001384292
possible label 0.0013716169999999999
unseen words 0.0013643050000000001
whole words 0.0013512420000000001
training examples 0.001319795
positive training 0.001315749
labeled training 0.001312819
sequence crf 0.001310897
training weights 0.001299059
same data 0.001296018
enough training 0.001270486
discriminatively training 0.001270486
segmentation models 0.001217894
form models 0.001213176
beijing corpus 0.001193221
generative models 0.001182132
character label 0.001176609
test set 0.001175588
crf segmentation 0.001144525
other labels 0.001141022
markov crf 0.0011245650000000001
discriminative models 0.001122011
