different language 0.002686842
same language 0.002416746
source language 0.002402122
training language 0.002392174
test language 0.002274593
target language 0.002161028
language third 0.002030143
language models 0.002015856
related language 0.002001806
lated language 0.0020010600000000003
finnish language 0.001988936
english corpus 0.001974509
distant language 0.001965124
language families 0.001960706
language spilling 0.001959287
indicated language 0.001959287
training corpus 0.0017724540000000001
machine translation 0.00174439
language 0.00173323
test data 0.001728413
test corpus 0.001654873
different source 0.0016225039999999999
translation quality 0.001570886
different languages 0.001550451
german corpus 0.00153559
different set 0.001484001
main corpus 0.001468406
large corpus 0.001458127
corpus construction 0.001448714
french corpus 0.001440983
english text 0.001431581
english texts 0.001426908
development corpus 0.001412911
second corpus 0.001396721
full corpus 0.001381247
europarl corpus 0.001374953
combined corpus 0.001370955
finnish corpus 0.0013692160000000001
original english 0.001368038
iht corpus 0.001366573
same source 0.001352408
comparable corpus 0.001349792
testing corpus 0.0013457550000000001
function words 0.001343234
roparl corpus 0.001340015
corpus hold 0.001340015
other source 0.001335705
translation 0.00131073
tion words 0.0013091679999999999
different domains 0.001283665
source languages 0.001265731
other languages 0.001263652
source text 0.001239474
different genres 0.001230386
training texts 0.001224853
content words 0.001210428
words frequencies 0.001197154
grammatical words 0.00117895
learning method 0.00117869
several source 0.001178669
underrepresented words 0.001176678
test corpora 0.0011665669999999999
english articles 0.00112439
corpus 0.00111351
english chunks 0.001111962
same target 0.001111314
english component 0.00109849
inal english 0.0010933779999999999
english this 0.001093234
same experiments 0.001081676
several text 0.001080359
same result 0.001075349
word freq 0.001075061
similar source 0.001072837
such features 0.001042258
such differences 0.001028056
corpora first 0.001027464
multiple source 0.00102468
same size 0.001024412
source lan 0.0010106100000000001
single source 0.0010025350000000001
target text 9.9838E-4
lexical features 9.88159E-4
spanish texts 9.87747E-4
ferent source 9.65818E-4
words 9.49896E-4
specific source 9.457160000000001E-4
other genres 9.435870000000001E-4
same pattern 9.41176E-4
related source 9.374680000000001E-4
same purpose 9.26567E-4
such effects 9.24929E-4
cognate languages 9.235370000000001E-4
high accuracy 9.225030000000001E-4
feature set 9.20372E-4
unrelated source 9.144960000000001E-4
same tar 9.1407E-4
source languag 8.95084E-4
fifth source 8.94307E-4
ferent languages 8.93765E-4
