unlabeled data 0.002796158
labeled data 0.002612044
training data 0.002585691
test data 0.002448702
data structure 0.002432785
lpjs data 0.002164717
data increases 0.002156876
word sense 0.002153072
evaluation data 0.00215266
benchmark data 0.002148481
unannotated data 0.002141963
data sparseness 0.002139712
classified data 0.002136972
unlabelled data 0.002134405
sified data 0.002134019
labled data 0.002134019
different sense 0.001915048
learning algorithm 0.0018879909999999999
feature set 0.001833802
sense tag 0.001717189
feature clustering 0.001710754
bootstrapping sense 0.001708109
sense cluster 0.001682197
sense disambiguation 0.001669754
sense tags 0.001658912
label matrix 0.001624703
feature selection 0.001616887
unlabeled examples 0.001606965
multiple sense 0.001598712
unsupervised feature 0.00159064
distance matrix 0.0015792459999999999
sense dis 0.001566189
bootstrapping algorithm 0.0015641189999999999
frequent sense 0.001557166
sense inventory 0.001544915
predefined sense 0.001544507
priate sense 0.001544028
sense heuristics 0.001544028
based algorithm 0.001538717
propagation algorithm 0.0015144059999999998
feature vec 0.001503782
feature prob 0.00148071
noisy feature 0.0014796850000000001
label information 0.00145932
labeled examples 0.001422851
supervised word 0.0014078609999999998
ping algorithm 0.001402725
learning method 0.001386215
labeled set 0.001350916
ing label 0.0013396950000000001
training set 0.001324563
word senses 0.001313358
other words 0.00130626
other examples 0.001304076
classification result 0.001295619
unlabeled points 0.0012927569999999998
supervised learning 0.00128677
sense 0.00128443
ing results 0.0012835099999999999
test examples 0.0012595089999999998
final classification 0.001242779
labeled dataset 0.001235166
connected graph 0.00122541
ambiguous word 0.001222809
feature 0.0012198
transition matrix 0.001217746
similar examples 0.001214228
only features 0.001211043
learning methods 0.0011946840000000001
test set 0.0011875739999999998
unlabeled exampleb 0.0011799509999999998
same class 0.001176844
model selection 0.001175432
learning algorithms 0.001157728
classification hyperplane 0.001153931
ideal classification 0.0011538680000000002
label entropy 0.001152336
classification goal 0.0011497830000000001
classification plane 0.001148302
classification operation 0.001148302
classification hyper 0.001148302
labeled example 0.001145568
many examples 0.0011423919999999999
algorithm 0.00114044
label propagation 0.001134639
matrix tij 0.001132824
identity matrix 0.001129225
splitting matrix 0.001125332
label sources 0.001109465
labeled points 0.0011086429999999999
average distance 0.00110649
soft label 0.001102432
similarity information 0.001100019
disambiguation results 0.001089812
english words 0.001082095
correct label 0.001081766
prediction results 0.001063679
target words 0.001058443
tag set 0.001046761
knn model 0.001038963
