sentiment model 0.00526917
sentiment word 0.00506208
sentiment words 0.00489693
twitter sentiment 0.004503248
sentiment classification 0.004392361
sentiment models 0.004388216
sentiment score 0.004227822
sentiment analysis 0.004216636
sentiment scores 0.004103857
sentiment classifier 0.004091746
new sentiment 0.004058134
sentiment labels 0.004048385
sentiment mixture 0.0040449859999999995
multiple sentiment 0.00401153
sentiment class 0.004007926
last sentiment 0.004006308
hashtag sentiment 0.003990845
current sentiment 0.003981793
separate sentiment 0.003974796
sentiment probabilities 0.003935269
universal sentiment 0.003918907
run sentiment 0.003917121
sentiment lexi 0.00390354
sentiment classes 0.00390189
independent sentiment 0.003898881
sentiment mix 0.0038981149999999997
sociated sentiment 0.0038971179999999998
sentiment 0.00365959
topic model 0.00297838
word lexicon 0.0022321
model parameter 0.002126002
twitter data 0.002059758
negative word 0.001997055
mixture model 0.001994976
lda model 0.00196488
ment model 0.001949081
model estimation 0.001936475
timent model 0.001926973
topic information 0.001913358
topic distribution 0.0019042310000000002
svm model 0.001888666
ture model 0.0018837910000000001
inference model 0.001871298
universal model 0.001868897
model outper 0.0018503600000000001
model avg 0.001847275
negative words 0.001831905
training data 0.00182406
negative data 0.001810665
positive words 0.001775565
top topic 0.0017658840000000001
ing data 0.0017342579999999998
lexicon features 0.0017073040000000002
negation word 0.001672057
topic modeling 0.001661652
same data 0.001650796
topic informa 0.001644735
topic inference 0.001630518
universal topic 0.001628117
topic distributions 0.0016273120000000001
model 0.00160958
topic distribu 0.001607647
likely topic 0.001606887
topic identification 0.001606887
topic iden 0.001606887
twitter domain 0.00159442
media data 0.00158793
testing data 0.0015432389999999999
additional data 0.0015373639999999998
tive words 0.001533371
data selection 0.001530472
frequent words 0.001527446
annotated data 0.001507996
negation words 0.001506907
informal words 0.001497421
ter data 0.001496863
stop words 0.001490331
elongated words 0.00148094
cluster data 0.001480908
newswire data 0.0014640259999999998
topic 0.0013688
classification accuracy 0.001276414
words 0.00123734
classification time 0.001224117
classification problem 0.001188416
support vector 0.001183514
third features 0.001171577
ment lexicon 0.001169111
lexical features 0.001164642
feature space 0.001164115
negative score 0.0011627970000000001
vector machine 0.001155887
twitter nlp 0.001154525
contextual features 0.001151757
various features 0.001135704
local features 0.001133398
corresponding feature 0.001122589
positive score 0.001106457
generative models 0.001093992
manual lexicon 0.001090433
