feature set 0.002512264
language model 0.00246546
classification features 0.0023745899999999998
syntactic features 0.002331312
feature approach 0.002280337
tsg feature 0.002274382
cfg feature 0.0022444229999999997
tsg features 0.0022148719999999997
ing features 0.0021489969999999997
binary feature 0.002091037
reranking features 0.0020673849999999997
discriminative features 0.002059605
latent feature 0.002047179
feature sets 0.002033936
btsg feature 0.002004258
feature func 0.001999461
feature redundancy 0.0019954309999999998
superior features 0.0019789029999999997
induced features 0.001974244
correlated features 0.001935968
cfg model 0.001818973
training set 0.001814644
model accuracy 0.001785418
ing model 0.001783057
feature 0.00175878
tree fragments 0.001754714
performance model 0.0017414169999999999
bayesian model 0.001725692
such tree 0.001719006
features 0.00169927
training data 0.0016775750000000002
induction model 0.001668587
regression model 0.001660443
cfg language 0.001617773
parse tree 0.001612613
model voteone 0.0015696339999999999
native language 0.001522551
grammars tree 0.001493616
tree kernel 0.0014343519999999998
language detection 0.001433995
test set 0.001407807
ferent language 0.001406081
tree substitution 0.001401081
tive language 0.001400056
data set 0.001369899
model 0.00133333
tree structure 0.001316414
tree kernels 0.001314387
parsing results 0.001302607
elementary tree 0.001294232
large set 0.001283634
tsg fragments 0.001273804
different classification 0.001273087
tree substitu 0.001232993
substitution grammar 0.001226516
sampling algorithm 0.001224912
set size 0.001185784
related work 0.0011773719999999999
grammar induction 0.001157204
split grammar 0.001154381
other methods 0.001136719
language 0.00113213
classification accuracy 0.0011274079999999999
single fragment 0.001123584
different tsg 0.001113369
full set 0.001090342
tsg rules 0.0010895850000000001
classification tasks 0.001087695
small fragments 0.001084419
classification performance 0.001083407
detection work 0.0010807619999999999
unsupervised grammar 0.00108053
tsg parsing 0.001078912
grammar sizes 0.001066149
statistical parser 0.001061196
training 0.00106116
finite grammar 0.001061102
grammar induc 0.001060087
berkeley parser 0.0010599099999999998
cfg rules 0.0010596260000000001
manageable grammar 0.00105936
shared fragments 0.001054504
text classification 0.001047003
multiple fragments 0.001039675
maximal fragments 0.001032326
large grammars 0.001027254
pling algorithm 0.001017653
parse trees 0.001016873
set sizes 9.97686E-4
word list 9.95113E-4
large number 9.94352E-4
eter set 9.939670000000001E-4
art results 9.93131E-4
bayesian models 9.92722E-4
question classification 9.875959999999999E-4
fragment occurrence 9.85199E-4
noisy results 9.82549E-4
english text 9.74747E-4
based classification 9.74376E-4
maximum probability 9.65269E-4
