clustering model 0.002005574
language model 0.001896788
word similarity 0.0018592300000000001
cluster model 0.001806842
test data 0.001724906
training data 0.0016763759999999998
ing data 0.001632662
new data 0.0015338539999999999
new model 0.001529724
data problem 0.001466775
model size 0.001440831
data reduction 0.001429955
probabilistic model 0.00140898
model complexity 0.00139511
single model 0.001392915
clustering performance 0.0013804540000000001
data com 0.001379204
data variables 0.0013716289999999999
model prob 0.001367579
rent model 0.001356111
dividual word 0.001353295
different probability 0.00133892
sparse data 0.001333183
actual data 0.001329596
data preparation 0.001328326
bility model 0.00132671
data sparseness 0.001325712
lier model 0.001321439
similar words 0.001306129
clustering results 0.00128913
clustering methods 0.001251072
cluster method 0.0012288770000000002
average clustering 0.001220322
distributional clustering 0.001204295
clustering case 0.001169548
similarity information 0.001169273
probability distribution 0.001160107
tional clustering 0.0011508920000000001
model 0.00113631
same cluster 0.001134702
ter words 0.001108679
ing test 0.001076688
cluster distribution 0.00107514
specific words 0.0010733840000000001
butional clustering 0.001065544
abilistic clustering 0.001063832
language modeling 0.001063121
tributional clustering 0.001057289
boolean clustering 0.001054621
clustering gener 0.001054621
ing method 0.001050567
distant words 0.001044925
probability estimation 0.001035007
probability estimates 0.001030888
distributional similarity 0.001026791
tant words 0.0010166560000000001
uninformative words 0.001015665
conditional probability 0.001011019
test set 0.0010042789999999998
cooccurrence probability 9.976479999999999E-4
similar performance 9.8795E-4
language mod 9.616539999999999E-4
zero probability 9.60003E-4
probability esti 9.45623E-4
test baseline 9.43983E-4
cooeeurrence probability 9.41218E-4
such estimates 9.294980000000001E-4
cluster membership 9.077110000000001E-4
previous work 8.88969E-4
cluster centroids 8.86918E-4
same set 8.839830000000001E-4
similarity mea 8.78541E-4
similarity infor 8.778880000000001E-4
different values 8.763180000000001E-4
own cluster 8.7436E-4
baseline performance 8.707070000000001E-4
original cluster 8.7018E-4
clustering 8.69264E-4
distributional models 8.66747E-4
aged cluster 8.56642E-4
entire test 8.482259999999999E-4
test cooccurrences 8.45691E-4
test type 8.404759999999999E-4
test sets 8.30716E-4
words 8.29369E-4
modeling performance 8.13833E-4
unseen pair 8.016309999999999E-4
different partitions 8.00673E-4
test triples 8.006179999999999E-4
different centroids 7.99807E-4
ing sequence 7.96988E-4
method yields 7.82356E-4
test instance 7.797079999999999E-4
test instances 7.733659999999999E-4
test fold 7.7228E-4
overall performance 7.72205E-4
different cooc 7.72156E-4
other tasks 7.694970000000001E-4
language 7.60478E-4
unseen cooccurrences 7.60239E-4
