word alignment 0.00816563
english word 0.0062786199999999995
word segmentation 0.0061794599999999995
word alignments 0.006115483
chinese word 0.006011272999999999
same word 0.005955841
different word 0.005706007999999999
statistical word 0.005705541999999999
initial word 0.0056106839999999995
word level 0.005583166
word align 0.005578255999999999
automatic word 0.005543083999999999
word segmenter 0.0055414209999999995
ibm word 0.0055175599999999995
word segmenta 0.005505723
single word 0.005493688
ing word 0.0054831839999999995
word aligner 0.005455326999999999
word boundaries 0.005442313999999999
word aligners 0.005428774
word sequences 0.005426452
word packing 0.005424628
word repacking 0.005423753
matic word 0.005412792
gle word 0.0054113799999999995
bootstrapped word 0.005410162
deterministic word 0.005410162
alignment model 0.00406659
bilingual alignment 0.003895038
training alignment 0.003866289
statistical alignment 0.003623312
alignment models 0.003620561
alignment error 0.0034892020000000003
alignment process 0.003435809
alignment types 0.00342882
alignment output 0.003390692
alignment candidates 0.003338602
reliable alignment 0.003335777
alignment confi 0.003327762
alignment 0.0030417
english words 0.0027437299999999998
translation system 0.002717224
other words 0.002440373
translation task 0.0023142989999999997
english corpus 0.00231085
machine translation 0.002276872
translation quality 0.00212104
target words 0.002104769
english sentence 0.002103962
translation output 0.002099152
initial words 0.0020757939999999997
final translation 0.00204772
bilingual corpus 0.002009498
training corpus 0.001980749
chinese segmentation 0.001942873
eral words 0.001923848
consecutive words 0.001905813
chinese sentence 0.001836615
same method 0.001762034
translation 0.00175016
language model 0.001717514
english sentences 0.001690511
first english 0.0016697640000000002
smt system 0.00161605
additional segmentation 0.0015996220000000002
words 0.00158904
iwslt english 0.001576474
various corpus 0.001558162
english dictionary 0.0015579500000000002
parallel corpus 0.001544023
other language 0.0015439569999999999
training data 0.001535261
few english 0.001534378
english side 0.001529154
training set 0.001526861
english train 0.0015202260000000002
expression corpus 0.001512937
speech corpus 0.001512537
new alignments 0.001504303
sentence pair 0.0014955839999999999
corpus statistics 0.0014901900000000002
automatic segmentation 0.001474684
europarl corpus 0.001474291
europarl english 0.001472821
new sentence 0.0014620219999999999
baseline system 0.0014542679999999999
english reference 0.001444458
corpus accord 0.0014431140000000001
dated corpus 0.0014431140000000001
english datasets 0.0014418
ibm model 0.0014185200000000002
sentence pairs 0.0014040670000000002
simple method 0.00139186
test set 0.001382915
original segmentation 0.0013691670000000001
automatic method 0.001349277
segmentation configuration 0.001346025
segmentation strategy 0.001344371
output alignments 0.001340545
original model 0.0013385270000000002
