training sentences 0.002839406
training data 0.0027337530000000002
large training 0.002579664
training set 0.002552858
selection training 0.00255104
training corpus 0.00251787
training examples 0.0024394490000000002
small training 0.0024061760000000003
high training 0.002314919
training process 0.002294986
labeled training 0.002288156
training corpora 0.002274434
supervised training 0.002262819
annotated training 0.002236112
training candidates 0.002218851
training exam 0.0022095780000000002
training brackets 0.002204179
training annotations 0.002197106
training pool 0.002192971
helpful training 0.002183168
training utility 0.002174215
select training 0.002170588
training utilities 0.002168104
training iteration 0.002167762
building training 0.002166875
uulabeled training 0.002166875
training loop 0.002166875
training supervision 0.002166875
ceive training 0.002166875
training pools 0.002166875
parse tree 0.001936755
training 0.00190353
tree entropy 0.001811597
sentence length 0.001809948
test sentences 0.001749422
learning algorithm 0.0017075940000000002
test data 0.001643769
selection learning 0.001634657
different set 0.001594951
tree structure 0.001542551
new sentences 0.0015166329999999999
possible sentences 0.001488199
tree ent 0.0014251
likely tree 0.001421884
other function 0.001414116
lexicalized tree 0.0014087470000000001
learning algorithms 0.001392222
candidate sentence 0.001390342
tree insertion 0.001388509
test score 0.001388
tened tree 0.001384711
learning process 0.001378603
learning system 0.0013783600000000001
length evaluation 0.0013770800000000001
possible parse 0.001368308
selection algorithm 0.001367957
learning rate 0.001357637
sentence lengths 0.001352241
learning problem 0.001344217
unlabeled sentences 0.001330553
parse trees 0.001327876
evaluation function 0.0013271770000000001
learning systems 0.001325056
probability distribution 0.001323528
ropy sentence 0.00131933
wsj sentences 0.001316218
ing sentences 0.001310309
simplistic sentence 0.001310143
entropy evaluation 0.001304009
syntactic parse 0.00129758
learning algo 0.001291667
candidate sentences 0.001280168
learning rates 0.0012769860000000001
anguage learning 0.001270148
full sentences 0.00126409
lection learning 0.00126188
traditional learning 0.001251
grammar size 0.001247503
different sizes 0.0012240620000000002
different hypothesis 0.001222005
different aspects 0.001219145
labeled data 0.001214849
correct parse 0.001206619
ing data 0.0012046560000000001
new corpus 0.001195097
sample selection 0.001185453
test sen 0.001175923
grammar induction 0.001160291
small set 0.001151974
annotated parse 0.001148567
induction algorithm 0.001146993
potential parse 0.00113619
particular parse 0.001135871
entropy value 0.00113056
available data 0.001127191
parsing accuracy 0.001117716
likely parse 0.0011170989999999999
new examples 0.001116676
stochastic grammar 0.001115431
entropy definition 0.00111315
