word alignment 0.00943339
word translation 0.008300160000000001
word alignments 0.00754377
data word 0.00753233
bilingual word 0.00750778
word collocation 0.00745932
monolingual word 0.007243783
word pair 0.007236660000000001
word pairs 0.007174375
target word 0.007000704
word sequence 0.006990925
word align 0.006978539000000001
word distortion 0.006975496
word colloca 0.006933642
single word 0.006919245
lingual word 0.006895513
word collo 0.006875504
ferent word 0.006861794
consecutive word 0.00686034
word units 0.0068445500000000005
gual word 0.006842936
improved word 0.00683979
directional word 0.006831107
proved word 0.0068302810000000005
interrupted word 0.006828969
thg word 0.006828145
rupted word 0.006828145
thk word 0.006828145
alignment method 0.004299320000000001
alignment set 0.003816743
same alignment 0.003795654
alignment links 0.0037393030000000002
alignment models 0.003631815
alignment results 0.0036080030000000003
alignment improvement 0.003542612
alignment algorithm 0.003526304
alignment error 0.003507645
alignment sequence 0.003477935
alignment baseline 0.003477593
baseline alignment 0.003477593
alignment probabilities 0.00346794
alignment process 0.003410442
optimal alignment 0.003388324
alignment probabili 0.003333429
alignment union 0.003327452
directional alignment 0.003318117
alignment perfor 0.003316515
translation model 0.00330414
alignment 0.0029602
translation probability 0.0027221470000000003
bilingual corpus 0.0026070499999999996
language model 0.00257271
model collocation 0.0024633
collocation model 0.0024633
alignments method 0.0024097
monolingual corpus 0.002343053
translation probabilities 0.0023347100000000003
translation quality 0.002278459
words english 0.002248046
english words 0.002248046
training method 0.002226947
chine translation 0.002183519
chinese words 0.002154728
bilingual sentence 0.00210413
parallel corpus 0.002073928
fertility model 0.002002722
lingual corpus 0.001994783
source words 0.001987152
distortion model 0.001979476
ibm model 0.001960931
fbis corpus 0.001958087
training data 0.001946967
same language 0.0019309940000000001
distribution model 0.001929128
norama corpus 0.0019277729999999998
mwa model 0.00191114
tion model 0.0018912590000000002
scfg model 0.0018900660000000001
collocation probability 0.001881307
bilingual phrase 0.001879221
bwa model 0.0018608890000000001
baseline method 0.0018565130000000002
different collocation 0.0018444400000000001
monolingual sentence 0.0018401329999999999
sentence pair 0.00183301
tribution model 0.001832333
phrase collocation 0.001830761
translation 0.00182697
mwa method 0.0017730900000000002
sentence pairs 0.001770725
source language 0.0017458320000000001
smt system 0.0017392549999999999
novel method 0.001734857
bwa method 0.001722839
source sentence 0.001719832
unrelated words 0.001700194
collocated words 0.00169622
symmetrization method 0.0016949740000000001
irrelevant words 0.0016926760000000002
monolingual collocations 0.001660282
