parallel data 0.0028884699999999998
chinese data 0.002781773
english data 0.002767943
sentiment data 0.002708093
word translation 0.0027066599999999996
labeled data 0.002644947
training data 0.002615705
unlabeled data 0.002596294
test data 0.002558493
same data 0.002509336
data set 0.002431782
data table 0.00237639
language words 0.002371546
language sentence 0.002345987
classification model 0.002340454
data sets 0.002326435
task data 0.002326343
chinese translation 0.0022986029999999998
english translation 0.002284773
other language 0.0022752859999999996
target language 0.002254293
chinese word 0.002197063
english word 0.0021832329999999997
multiple data 0.0021767519999999997
data clmm 0.002176309
words translation 0.002166926
allel data 0.002144486
ntcir data 0.002138001
source language 0.002095143
machine translation 0.002011728
model figure 0.001854136
language gap 0.001853249
language wordsws 0.0018511579999999999
model parameters 0.001841776
word alignment 0.00183818
model use 0.0017886360000000001
generative model 0.001771638
mixture model 0.001739448
sentiment classification 0.001732327
clmm model 0.001717989
chinese sentiment 0.001715326
ture model 0.001710872
unlabeled parallel 0.0017102240000000002
english sentiment 0.001701496
parallel corpus 0.00167682
translation engines 0.001665052
use word 0.001662246
chinese labeled 0.00165218
translation oracle 0.001646139
word projection 0.001638497
english labeled 0.00163835
chinese sentence 0.00163177
english sentence 0.00161794
language 0.00160872
sentiment words 0.001583649
labeled sentiment 0.0015785
word align 0.001573562
word alignments 0.001560211
word projec 0.0015558579999999998
parallel sentences 0.001533923
same sentiment 0.0014428890000000002
model 0.00142895
chinese sentences 0.0014272260000000002
bilingual parallel 0.001418888
chinese classifier 0.001417639
english sentences 0.001413396
classification accuracy 0.00141333
translation 0.0014041
labeled text 0.001400254
machine learning 0.0013916900000000001
english documents 0.001354156
classification task 0.001350577
sentiment classifier 0.001343959
sentiment analysis 0.001337637
labeled sentences 0.0012904000000000001
chinese classifiers 0.001287553
parallel cor 0.0012759210000000002
training set 0.001272947
unlabeled sentences 0.001241747
training process 0.0012284560000000002
different set 0.0012283530000000002
english product 0.001218938
learning models 0.0012156649999999999
many sentiment 0.001202464
ment classification 0.0011914590000000002
negative english 0.001188112
chinese sen 0.00118158
timent classification 0.001166593
only english 0.001163643
ing methods 0.001159051
chinese docu 0.001158946
conventional classification 0.001157421
classification settings 0.001155119
classification accu 0.0011539900000000001
sentiment polarity 0.00114938
monolingual sentiment 0.001136229
source target 0.001131996
sentiment scores 0.0011239450000000001
english andchinese 0.001123372
binary sentiment 0.00110196
