sentiment model 0.0051607300000000005
sentiment features 0.004870622
sentiment lexicon 0.004804813
sentiment label 0.004750595
negative sentiment 0.004668703
sentiment models 0.004661248
positive sentiment 0.0046434480000000005
sentiment polarity 0.004605907
sentiment analysis 0.004561191
sentiment table 0.004497788
sentiment entity 0.004463659
sentiment target 0.004435096
sentiment prediction 0.004428932
sentiment task 0.0044144760000000005
prior sentiment 0.004383351000000001
joint sentiment 0.004378665
overall sentiment 0.004372361
sentiment baseline 0.004359104
sentiment lexicons 0.0043515310000000005
values sentiment 0.004335424
sentiment values 0.004335424
neutral sentiment 0.0043140620000000005
sentiment labels 0.004303836
sentiment sequence 0.004276530000000001
message sentiment 0.004257924000000001
modeling sentiment 0.004252161
sentiment base 0.004246675
sentiment precision 0.004243694
sentiment targets 0.004243651
sentiment span 0.004241964
sentiment variables 0.004239001
sentiment baselines 0.0042303620000000005
tive sentiment 0.0042214110000000004
sentiment recall 0.004217931
targeted sentiment 0.004209166
itive sentiment 0.004207519
sentiment lex 0.004203058
sentiment pre 0.004196801
sentiment instances 0.004193405
sentiment detec 0.004191155
public sentiment 0.004190412
sentiment tar 0.004189574000000001
sentiment percentages 0.004188548
geted sentiment 0.004188548
sentiment irre 0.004188548
sentiment val 0.004188548
sentiment 0.00395339
language model 0.0019304270000000002
lexicon word 0.001866903
features features 0.001834464
opinion words 0.0016326599999999998
model joint 0.001632615
joint model 0.001632615
crf model 0.001620656
english model 0.001578329
language data 0.0015693159999999999
word table 0.001559878
second model 0.0015516640000000001
first lexicon 0.001493968
basic model 0.0014871930000000002
graphical model 0.0014861030000000001
target words 0.0014800809999999998
collapsed model 0.001453435
model strategies 0.001442972
function word 0.001438638
function words 0.001421533
word tags 0.00141526
training data 0.001410873
linguistic features 0.001387117
word length 0.001377479
negative polarity 0.00136783
data twitter 0.0013515139999999998
entity label 0.001307474
pos features 0.0013004660000000001
word forms 0.001290778
data set 0.001290011
training models 0.001272502
word identity 0.0012663000000000002
unsupervised word 0.001260225
curse word 0.001258833
word clusters 0.0012569970000000001
abbreviation word 0.001253372
word lengthening 0.001251521
slang words 0.001244952
curse words 0.0012417279999999999
identical words 0.0012356799999999999
laugh words 0.001234254
spanish lexicon 0.001231741
further features 0.0012225600000000001
first work 0.001220587
clustering features 0.001212835
model 0.00120734
weighted features 0.001203041
only features 0.001202191
such models 0.001201588
surface features 0.001200962
polarity table 0.001196915
second lexicon 0.001195747
timent lexicon 0.001187815
features cluster 0.001178159
