sentiment rating 0.002857523
sentiment labels 0.002696949
sentiment score 0.0026609900000000002
sentiment prediction 0.002590683
sentiment analysis 0.002576519
positive sentiment 0.002567333
overall sentiment 0.002563564
sentiment classification 0.002541755
level sentiment 0.002515239
other model 0.002456325
sentiment summarization 0.002434113
rating model 0.002407293
automatic sentiment 0.002392838
actual sentiment 0.002385574
predicted sentiment 0.002382838
regression model 0.00224237
learning model 0.002228061
linear model 0.002209772
sentiment 0.00216766
model figure 0.00209897
new model 0.002094188
mir model 0.002037219
initial model 0.001999527
model complexity 0.001974636
model settings 0.001966145
sion model 0.00195546
talks model 0.001954302
mixture model 0.001946304
games model 0.001931079
word vector 0.0018782539999999998
model 0.00171743
aspect rating 0.001631633
word vectors 0.001534161
aspect labels 0.0014710589999999998
other words 0.001455205
aspect ratings 0.001384197
word fea 0.001381817
aspect class 0.001379831
word pres 0.0013269159999999998
particular aspect 0.001291783
vector regression 0.0012869539999999999
other models 0.001284192
specific aspect 0.001272409
different feature 0.0012601589999999998
joint aspect 0.001236848
training data 0.001216365
instance feature 0.001208523
aspect identification 0.001186932
rating values 0.001174902
feature learning 0.001166746
dimension aspect 0.0011642549999999999
aspect rat 0.0011636469999999999
aspect detec 0.001152057
analysis data 0.001129474
rating class 0.001127924
text regression 0.001121023
rating prediction 0.001112886
other methods 0.0010937540000000002
topic models 0.001093246
instance regression 0.001077348
single vector 0.001074174
feature vectors 0.001074036
regression models 0.001070237
instance learning 0.001063039
average rating 0.001061825
data point 0.001059986
rating datasets 0.001051022
other baselines 0.001042343
linear models 0.001037639
rating inference 0.001036265
feature space 0.0010337229999999998
input data 0.001024203
instance algorithm 0.001021718
other nlp 0.001018061
linear regression 0.001017282
previous models 0.001015651
training set 0.001013955
class label 0.001011917
data sets 0.001009786
top words 0.001004473
other studies 0.00100169
emotion rating 9.92949E-4
other products 9.91569E-4
instance weights 9.90496E-4
other comments 9.89373E-4
joint rating 9.84941E-4
other combinations 9.80589E-4
port vector 9.73804E-4
test data 9.69163E-4
other side 9.69019E-4
class labels 9.6735E-4
ψˆi distribution 9.625009999999999E-4
maximum rating 9.62358E-4
various feature 9.57304E-4
same time 9.54852E-4
entire data 9.48228E-4
feature sets 9.45286E-4
aspect 9.4177E-4
pect rating 9.3968E-4
ing algorithm 9.393E-4
