domain feature 0.0037351100000000003
domain sentiment 0.0034690800000000003
word feature 0.0031842
target domain 0.002910447
domain document 0.0028989470000000002
source domain 0.00289267
word features 0.0027986499999999997
domain adaptation 0.002741835
work domain 0.0025285250000000002
domain kitchen 0.002525419
domain documents 0.002471158
cross domain 0.002459185
many domain 0.002449936
sentiment classification 0.002391218
domain divergence 0.002389495
domain pair 0.002388999
domain books 0.002374127
feature vector 0.002360539
feature representation 0.0023460110000000003
latent feature 0.0023306810000000002
domain doc 0.002291388
domain adap 0.002290851
parison domain 0.002288065
domain adapta 0.002286948
duct domain 0.002285907
domain projectors 0.002285907
dress domain 0.002285907
pivot feature 0.002209504
training data 0.002183081
labeled data 0.002164456
feature representations 0.0021557580000000002
feature space 0.002150491
new feature 0.002131643
word representation 0.002122891
latent word 0.002107561
original feature 0.002106078
feature mapping 0.002037157
domain 0.00203145
dimensional feature 0.0020085750000000003
supervised word 0.0020075830000000003
classification model 0.001993698
feature selection 0.001986808
tent feature 0.001978479
heuristic feature 0.0019725940000000003
generalizable feature 0.001971876
word cluster 0.001964299
feature replication 0.001960796
eralizable feature 0.001959167
latent features 0.001945131
learning method 0.001944503
word representations 0.001932638
data labels 0.0019180500000000001
word variables 0.001889264
original word 0.0018829580000000001
representation learning 0.001880311
unlabeled data 0.001871384
test data 0.0018545620000000001
word distributions 0.001852947
word clustering 0.0018408180000000001
input word 0.00183728
target classification 0.0018325849999999999
conditional word 0.0018309160000000001
pivot features 0.001823954
negative sentiment 0.001823725
learning methods 0.001822232
unsupervised word 0.001812526
word clusters 0.0018063670000000001
positive sentiment 0.0018060250000000002
model training 0.0017938809999999998
word topics 0.001789818
input data 0.00178605
soft word 0.0017678070000000001
supervised learning 0.001765003
hard word 0.001756187
tent word 0.001755359
word distri 0.001750258
word clusterings 0.001739998
pervised word 0.001739185
data annotations 0.00173749
sentiment polarity 0.001737157
ingful word 0.001735746
word clus 0.001735746
data sets 0.001733722
sentiment classifi 0.001733512
sentiment clas 0.0017322680000000001
original features 0.0017205279999999998
sentiment classifica 0.0017038130000000002
feature 0.00170366
reuters data 0.001700984
data zts 0.001687759
data ztr 0.0016866890000000001
data xts 0.001683957
specific features 0.0016759359999999998
latent model 0.0016671309999999999
classification results 0.001644842
ing features 0.0016426519999999999
correspondence learning 0.001640033
learning system 0.0016190549999999999
source training 0.001614991
labeled target 0.001614143
