phrase translation 0.0070665699999999994
target phrase 0.00486608
source phrase 0.00484304
phrase table 0.004618283
translation model 0.00457958
words phrase 0.004578301
large phrase 0.004482163
translation system 0.0044562099999999995
pivot phrase 0.004360663
phrase pairs 0.004199492
wrong phrase 0.004131519
phrase senses 0.004125581
final phrase 0.004123906
phrase tables 0.004109405
german phrase 0.004082218
irrelevant phrase 0.004074803
combined phrase 0.0040706200000000005
translation time 0.003980649
phrase 0.00377961
translation systems 0.003772135
machine translation 0.0037357519999999997
translation pairs 0.0037068419999999997
good translation 0.003675291
translation process 0.003654695
wrong translation 0.003638869
final translation 0.003631256
translation stage 0.003626612
translation correct 0.0036039089999999998
correct translation 0.0036039089999999998
translation errors 0.0035861699999999996
translation vorschlagen 0.0035772069999999998
translation 0.00328696
target phrases 0.00290625
word alignments 0.002631136
language model 0.00240486
pivot phrases 0.002400833
training data 0.00235313
new phrases 0.002290679
target system 0.00225572
target language 0.00219871
parallel data 0.002185233
source language 0.00217567
unique phrases 0.002156119
phrases base 0.002149724
weak phrases 0.002127403
mapping phrases 0.002117843
middle phrases 0.002111121
parallel system 0.002075853
table system 0.002007923
target sentence 0.001989103
source sentence 0.001966063
system bleu 0.001960889
target translations 0.001957328
target context 0.001877883
first system 0.0018772799999999998
pruning method 0.001856654
source context 0.0018548430000000001
english language 0.00184749
phrases 0.00181978
training time 0.0017681889999999999
corpus system 0.0017350339999999999
other language 0.0017265610000000002
similar target 0.001698756
pivot language 0.001693293
parallel method 0.001681267
training corpus 0.0016402839999999999
language approach 0.001628918
system performance 0.0016029899999999999
second system 0.0016023349999999999
same language 0.0015744069999999999
target side 0.001572901
target languages 0.001567417
pseudo system 0.001557363
source side 0.001549861
pruning step 0.001546927
target corpora 0.0015317969999999999
source corpora 0.001508757
good language 0.001500571
base system 0.001499194
bridge system 0.0014965909999999998
language vectors 0.001488855
resource language 0.001485634
other translations 0.001485179
parallel corpus 0.001472387
farsi system 0.0014644789999999999
target vectors 0.001463085
corresponding source 0.001454429
source vectors 0.001440045
bridge language 0.001439581
many translations 0.001432422
other words 0.001413012
intermediate language 0.00140594
unique source 0.001399769
different languages 0.001396498
parallel systems 0.001391778
pivot context 0.0013724660000000001
enough training 0.001372395
training samples 0.0013648879999999999
same sentence 0.0013648
source dictionaries 0.001354397
