training data 0.00354291
english data 0.002819276
training error 0.0027775400000000002
data preposition 0.002710241
evaluation data 0.002479595
examples data 0.0024469889999999997
learner data 0.0024423099999999996
esl data 0.0024042039999999996
linguistic data 0.0023772579999999997
annotated data 0.0023214069999999997
native data 0.002314793
chinese data 0.0022866529999999996
data size 0.0022838999999999997
speaker data 0.002256203
spanish data 0.0022531609999999996
combined data 0.002252401
error language 0.00222589
language model 0.0022225500000000002
data negall 0.002220611
data consortium 0.002202328
correction system 0.002173898
error correction 0.002096438
training examples 0.002063279
text correction 0.001977238
english text 0.001934706
native training 0.001931083
preposition errors 0.001898851
training approaches 0.001881946
other language 0.0018753139999999999
negall training 0.001836901
training conditions 0.001829292
typical training 0.001827392
training paradigm 0.001826942
training setting 0.001821871
candidate word 0.001821774
training paradigms 0.001818317
training condition 0.001818317
tween training 0.001818317
training option 0.001818317
source system 0.001798887
baseline system 0.0017489600000000001
spelling errors 0.0016657
similar error 0.001665206
error rate 0.001648886
preposition correction 0.001645429
tree model 0.001644842
candidate correction 0.0015976319999999999
training 0.0015796
correction systems 0.001573633
this system 0.001570243
entropy model 0.0015577800000000001
system acc 0.0015526440000000002
confusion errors 0.0015522280000000001
source language 0.001551437
incorrect error 0.001549279
other candidate 0.0015464979999999999
line system 0.001538106
error rates 0.0015373890000000001
other systems 0.001522499
hybrid system 0.0015151440000000002
ing word 0.0015138880000000001
tion errors 0.001512093
article errors 0.0015110610000000002
column error 0.001504531
language research 0.0014985519999999998
errors rates 0.0014913690000000002
error patterns 0.001490618
guage model 0.0014784660000000001
correction task 0.001474162
word selection 0.001473516
natural language 0.001473325
first language 0.001468407
artificial errors 0.0014621220000000002
target word 0.001456456
error statistics 0.001442827
error statis 0.001436572
word sense 0.0014344450000000001
correction problem 0.001434183
native text 0.001430223
spelling correction 0.0014122779999999999
candidate set 0.001393469
able word 0.001377317
second language 0.001362631
correction models 0.001339571
learner english 0.001334966
text intro 0.0013181209999999998
language applications 0.001313449
large corpus 0.001308279
language groups 0.0013064729999999998
correct preposition 0.001304718
system 0.0012754
language backgrounds 0.001274391
source preposition 0.0012704180000000002
language preps 0.001266836
article correction 0.001257639
valid words 0.0012456
same set 0.0012218279999999999
candidate corrections 0.001218611
native english 0.001207449
error 0.00119794
