word sense 0.0035476199999999996
different word 0.003360264
word cluster 0.003353375
target word 0.003109416
word clustering 0.0030986819999999997
based word 0.003003653
source word 0.002960478
word level 0.002906927
word clusters 0.0028938839999999998
word sparsity 0.002693108
superior word 0.002675438
word sequences 0.002656624
word clusterings 0.0026521509999999997
word alignments 0.002650463
word tokens 0.002647026
word wsak 0.002644337
sense features 0.0025692700000000002
feature set 0.002513958
cluster features 0.0023750250000000002
feature vector 0.002346865
feature size 0.002254302
additional feature 0.002155193
feature engineering 0.002071359
expensive feature 0.002071163
target words 0.002043186
based features 0.002025303
labeled data 0.001945809
training data 0.001938043
different cluster 0.001907239
semantic features 0.001878108
same sense 0.0018490099999999999
words senses 0.001844553
feature 0.00182428
classification model 0.0018115800000000001
target language 0.001801336
good features 0.001790544
similar words 0.001767143
related words 0.001753748
test data 0.001746199
data size 0.001738592
monolingual data 0.001709716
effective features 0.001707924
language corpus 0.001672979
different clustering 0.001652546
source language 0.001652398
wordnet sense 0.001623914
mar words 0.00161619
unlabelled data 0.001600424
data sparsity 0.001598478
annotated data 0.001592983
marathi data 0.00159012
stop words 0.001579834
words wtk 0.001579153
data samples 0.001556303
data variation 0.0015556699999999999
data pool 0.001549411
sense annotation 0.0015468539999999999
sense disambiguation 0.001526623
language models 0.001491641
sentiment classification 0.001482832
different number 0.001479837
different senses 0.001464647
same classification 0.00145875
sense identifiers 0.0014569539999999999
cluster level 0.001453902
sense mark 0.001444122
features 0.00142485
target domain 0.001424107
language pairs 0.00140436
sense input 0.001401241
same clustering 0.001400072
sense identifier 0.001391981
natural language 0.001390202
model parameters 0.0013855260000000002
cluster size 0.001380197
marathi language 0.00137667
sentiment analysis 0.001375153
language units 0.001373986
different approaches 0.001370716
rich language 0.0013657
direct cluster 0.0013572150000000002
different languages 0.001355737
language processing 0.001352581
language gap 0.001349173
classification results 0.001345344
words 0.00133697
language family 0.00133593
classification accuracy 0.0013302000000000001
sentiment classifier 0.0013264330000000001
different polarity 0.001319639
training set 0.001319151
classification performance 0.001298868
based clustering 0.001295935
domain corpus 0.00129575
cluster list 0.001289541
clustering algorithm 0.001278763
cluster identifiers 0.001262709
optimal cluster 0.001244048
naive cluster 0.001234525
level sentiment 0.001232399
