chinese word 0.002408087
word segmentation 0.002401579
heterogeneous word 0.002312541
current model 0.002034592
labeling model 0.002020358
word structures 0.002019903
style word 0.0019706990000000002
model diﬀerent 0.001970012
incompatible word 0.001963652
training data 0.001963227
neous word 0.001956222
word classiﬁcation 0.001951931
mapping function 0.001950974
discriminative model 0.001937145
model pdd 0.001919383
ﬁnal model 0.0019003689999999998
uniﬁed model 0.0018746779999999998
model parameters 0.00186184
state feature 0.001856725
merged model 0.0018548269999999999
our model 0.0018503439999999999
eﬀective model 0.0018320049999999998
proposed model 0.0018320049999999998
training algorithm 0.001815028
training method 0.0017983880000000002
method training 0.0017983880000000002
feature templates 0.001785918
transition feature 0.001779099
new features 0.00176402
annotation data 0.001721836
feature vector 0.001715355
related features 0.001697978
feature vectors 0.001673971
mapping label 0.001634235
training dataset 0.001624774
model 0.0015806
training set 0.001575529
chinese words 0.001574069
pos tagging 0.001530588
heterogeneous data 0.001525338
mapping sequence 0.001521317
segmentation label 0.0015068199999999999
heterogeneous pos 0.0015016320000000001
label sequence 0.001485744
sequence label 0.001485744
score function 0.001473941
segmentation error 0.001462914
pos tag 0.001448783
ppd training 0.001447881
training datasets 0.001440747
heterogeneous annotation 0.0014334
family function 0.001409066
pdd training 0.001395123
objective function 0.001389099
features 0.00138087
ping function 0.00137951
iterative training 0.001367781
annotation corpora 0.001362078
feature 0.00135681
joint chinese 0.001352318
online training 0.001349063
joint segmentation 0.0013458099999999998
ctb mapping 0.001323755
training stage 0.001319453
label set 0.0013185200000000001
training phase 0.001312313
heterogeneous sequence 0.001304864
ctb annotation 0.0013038
character sequence 0.0012955179999999998
pos tags 0.001285034
mapping labels 0.0012800440000000001
tation data 0.001235556
data size 0.001228344
corresponding mapping 0.001223506
loose mapping 0.0012174970000000001
diﬀerent annotation 0.0012043610000000001
mapping table 0.0011994290000000001
nese pos 0.00119239
data scale 0.001176216
adaptation method 0.00117126
output sequence 0.001169294
erogeneous data 0.001167816
heterogeneous corpora 0.00116558
single words 0.001164732
task learning 0.001162185
labeling models 0.001161605
annotation guide 0.001160458
mapping func 0.001153542
other corpus 0.0011529869999999998
sections words 0.0011449820000000001
ctb corpus 0.001140517
pos tagsets 0.001138965
input sequence 0.001126476
sequence labeling 0.001126171
chinese treebank 0.001124796
exact mapping 0.0011189680000000001
function 0.00111607
certain annotation 0.001110471
mapping relations 0.001109177
chinese language 0.001103134
