word sense 0.00439048
training word 0.00362947
test word 0.003335948
word senses 0.003267455
other word 0.003265701
target word 0.00323877
similar word 0.003210306
accuracy word 0.003173947
ing word 0.003153834
many word 0.00313384
various word 0.003099592
single word 0.00302802
different features 0.0027645029999999998
semantic features 0.002571624
feature vector 0.002387704
different words 0.0023813230000000003
baseline feature 0.002356883
topic features 0.002201577
new feature 0.002189903
feature selection 0.002158905
feature sets 0.002151908
theoretic feature 0.002131122
sense baseline 0.002129833
various features 0.002119612
sample feature 0.0021180500000000002
feature relevance 0.002115599
candidate features 0.002096067
line feature 0.002094924
sense disambiguation 0.002090587
ith feature 0.002087224
linguistic features 0.002086968
feature selec 0.002077117
stepwise feature 0.002075858
feature rel 0.002075858
typical features 0.002038871
nonzero features 0.0020259559999999998
sense dis 0.001885793
target words 0.00187561
frequent sense 0.001873866
similar words 0.001847146
feature 0.00184427
features 0.00179328
training data 0.001784056
selection model 0.001690215
svm model 0.00168085
model complex 0.0016769579999999999
ambiguous words 0.001663386
gaussian model 0.001645321
model complexity 0.001620858
sense 0.00161722
training set 0.001473093
different senses 0.001465418
words 0.0014101
model 0.00137558
accuracy models 0.001253735
different settings 0.001250807
labeled data 0.001224863
data points 0.001195458
ontonotes data 0.0011632279999999999
ing algorithm 0.001137665
retic models 0.001085611
similarity threshold 0.001072516
selection algorithm 0.001071726
prior information 0.001044845
information theoretic 0.001034049
wsd use 0.001031762
supervised wsd 0.001028717
information criterion 0.001025432
supervised learning 0.001025384
learning problems 0.001022171
similarity determination 0.001000011
tributional similarity 0.001000011
same probability 9.97924E-4
other senses 9.86636E-4
information theo 9.80882E-4
akaike information 9.80882E-4
information theory 9.80882E-4
transfer learning 9.642450000000001E-4
wsd techniques 9.58968E-4
learning settings 9.588590000000001E-4
learning techniques 9.55635E-4
baseline fea 9.33156E-4
similar senses 9.31241E-4
learning meth 9.26124E-4
learning priors 9.15676E-4
learning configuration 9.11611E-4
learning formula 9.11611E-4
other methods 9.070370000000001E-4
main problem 8.80608E-4
rich set 8.7268E-4
training 8.5621E-4
models 8.53048E-4
corresponding senses 8.52201E-4
selection method 8.468709999999999E-4
lexical sample 8.36793E-4
residual error 8.200200000000001E-4
total number 8.05816E-4
other hand 8.01259E-4
bernoulli distribution 7.98942E-4
first term 7.885500000000001E-4
