different feature 0.002594588
feature set 0.002526545
word senses 0.0024836290000000002
feature values 0.002412309
content word 0.002397349
feature space 0.002351311
pos feature 0.0023505180000000002
ambiguous word 0.002344737
particular feature 0.002298221
feature selection 0.002289801
ous word 0.002261485
feature sets 0.002257065
feature types 0.002250352
word chief 0.002249495
biguous word 0.002246705
word frequen 0.0022402109999999998
dimensional feature 0.002232638
feature mismatch 0.002224204
sification feature 0.002216774
feature 0.00199371
speech features 0.001810122
pos features 0.001717518
model form 0.0016684949999999999
data problem 0.001660739
data estimation 0.0016444790000000001
ions features 0.001625208
collocation features 0.0016199159999999999
line data 0.001608759
dissimilar features 0.001589203
interdependent features 0.001584724
cation features 0.001584724
identifiable features 0.001584724
learning algorithm 0.001575601
interest data 0.001535102
data sample 0.001514666
nlp data 0.0014819120000000002
complete data 0.0014728480000000001
content words 0.001405169
model parameters 0.001375865
features 0.00136071
bayes model 0.001356197
probabilistic model 0.001320829
complete model 0.001314908
model param 0.001310864
frequent words 0.001305519
individual words 0.001281499
tent words 0.001250137
ferent words 0.0012483009999999998
learning problem 0.001221135
ing algorithm 0.001207161
learning algorithms 0.0011991760000000001
learning experiments 0.001151513
sense distribution 0.001133917
unsupervised learning 0.001114675
little information 0.00110794
same class 0.0011038229999999999
supervised learning 0.00109293
model 0.00108216
same sentence 0.001068435
tle information 0.001067922
annotated text 0.001048549
sense value 0.001047253
maximization algorithm 0.00103524
words 0.0010247
challenging learning 0.0010239840000000001
different values 0.001019477
untagged text 9.99361E-4
algorithm particu 9.99029E-4
clustering algorithms 9.875650000000001E-4
particular sense 9.575339999999999E-4
different experiments 9.51895E-4
similarity matrix 9.515020000000001E-4
single sense 9.40971E-4
frequent sense 9.33842E-4
ing training 9.31855E-4
same opinion 9.07195E-4
sense definition 9.04658E-4
class distribution 9.02396E-4
majority sense 9.00632E-4
skewed sense 8.90772E-4
clustering methods 8.904410000000001E-4
sense distri 8.88756E-4
known sense 8.87399E-4
knowledge sources 8.79983E-4
natural distribution 8.76732E-4
sense inventory 8.76454E-4
average accuracy 8.74674E-4
possible values 8.741E-4
journal corpus 8.59727E-4
training examples 8.57834E-4
accuracy disambiguation 8.54651E-4
other parts 8.50939E-4
information 8.43651E-4
current estimate 8.38765E-4
natural language 8.3799E-4
current value 8.36548E-4
similarity analysis 8.35228E-4
ing clusters 8.336859999999999E-4
clustering algo 8.32597E-4
different ypes 8.26218E-4
