word set 0.004354704
correct word 0.0042441630000000004
target word 0.004239007
common word 0.004229924
semantic word 0.004223126000000001
distributional word 0.004213658
word confusion 0.004148318
confusion word 0.004148318
word pair 0.00413504
order word 0.004125315
frequency word 0.0041064330000000005
candidate word 0.004097745
word pairs 0.0040954500000000005
word sets 0.004077048
word usage 0.004055845000000001
frequent word 0.004048212
appropriate word 0.004030841
word list 0.004026526
word relatedness 0.0040096
confusable word 0.003989491
sion word 0.003973894
word choice 0.003971693
butional word 0.0039687590000000005
exact word 0.003963416
wrong word 0.003963332
dicative word 0.00396271
stop word 0.00396271
lexical features 0.002530667
target words 0.0024070769999999997
common words 0.002397994
parse features 0.002357446
lex features 0.0023184679999999997
confusion words 0.002316388
contextual features 0.002310169
features svm 0.002301794
count features 0.0022342409999999997
contextual words 0.002233259
par features 0.002226091
salient features 0.002212856
rare words 0.002210137
confused words 0.002158532
confusable words 0.0021575609999999997
salient words 0.002135946
filler words 0.00213337
able words 0.002132533
meaningful words 0.002131131
textual words 0.002131131
words amples 0.002131131
information feature 0.002090491
training data 0.002013649
features 0.00197674
words 0.00189983
test data 0.001835115
parse feature 0.001777936
feature name 0.0017166949999999999
feature values 0.001704989
training corpus 0.0016976299999999999
feature classes 0.001687525
feature example 0.001662579
par feature 0.0016465809999999998
svm model 0.0016343339999999999
relevant feature 0.001630294
language models 0.00159807
regression model 0.001580901
gram model 0.0015585439999999998
ing data 0.0015197449999999999
baseline system 0.0015126
large training 0.0014703860000000002
feature 0.00139723
serious data 0.00138138
development data 0.001380243
correction system 0.001365494
system combination 0.0013606480000000001
nyt data 0.001357202
data sparsity 0.001335439
different counts 0.0013215289999999999
different approaches 0.001319686
data sizes 0.0013195799999999999
model 0.00130928
new errors 0.001275248
learning systems 0.001256755
svm models 0.001251498
distribution models 0.001247899
spelling error 0.001247626
other confusion 0.001243083
different sizes 0.00121053
error correction 0.001206215
different filler 0.00120202
ture models 0.001197549
supervised learning 0.001186912
semantic information 0.0011846270000000002
combined system 0.001181966
distributional information 0.001175159
mal training 0.001166951
powerful models 0.001157802
size corpus 0.0011277639999999998
test cases 0.0011155129999999998
english learning 0.001109327
machine learning 0.001107832
other factors 0.001101638
