test data 0.0023445849999999997
training data 0.002288992
data set 0.002212996
ing data 0.002061257
different feature 0.001997942
error correction 0.001922736
data sets 0.0018891049999999999
feature set 0.0018719459999999998
development data 0.001862273
semantic features 0.001860361
data increases 0.00181164
proprietary data 0.001807626
velopment data 0.0018056349999999999
other error 0.001801787
word corpus 0.001796773
additional features 0.001735808
words corpus 0.001705123
lexical features 0.001691863
input features 0.001668772
baseline model 0.001650033
feature space 0.0016449049999999999
grammatical error 0.0016415129999999998
word baseline 0.001625503
additional feature 0.0015979179999999998
feature vectors 0.001587245
various feature 0.0015650059999999999
feature extraction 0.001564231
number error 0.0015592079999999999
important feature 0.001556533
feature sets 0.001548055
above feature 0.0015453449999999998
original feature 0.0015392739999999998
error correc 0.001498685
particular feature 0.0014985209999999998
feature functions 0.001494028
error tags 0.001489097
complete feature 0.0014739439999999998
aso model 0.001471058
feature spaces 0.001466181
cal error 0.001456872
gigaword model 0.001455583
matical error 0.0014453819999999998
articles model 0.001410943
language models 0.0014004389999999999
nucle model 0.00139704
model prec 0.001396542
prepositions model 0.001396528
test set 0.0013871209999999998
single model 0.0013859290000000002
particular word 0.001382721
features 0.00138207
word choice 0.001373704
head word 0.001363087
learner errors 0.001361878
problem training 0.001361312
aso system 0.0013330080000000001
training set 0.001331528
preposition errors 0.0013245750000000001
grammatical errors 0.001318874
article errors 0.001307564
different performance 0.001306865
same test 0.001297898
head words 0.001271437
learner text 0.001267254
learning algorithm 0.001264215
system flags 0.001262593
lion words 0.001258839
tion errors 0.001255981
same correction 0.001245689
feature 0.00124418
correction task 0.001233124
different classifiers 0.0012265660000000001
error 0.00121559
tree learning 0.001181803
model 0.00115291
preposition correction 0.0011387699999999999
language process 0.001122138
article correction 0.001121759
different domain 0.0011194780000000001
cle errors 0.001118859
substitution errors 0.001117198
second language 0.001109395
natural language 0.001081535
auxiliary problem 0.001080209
first set 0.001057817
other learner 0.001055124
learning gec 0.0010461490000000001
artificial test 0.001038057
words 0.00103673
classification problem 0.001031325
different lin 0.00102804
different fea 0.001021062
tion problem 0.00102058
system 0.00101486
test instances 0.001011786
training examples 0.001010478
target problem 0.001007872
gigaword training 0.001006435
nucle test 0.001003485
test instance 0.0010018639999999999
