language word 0.002559909
english word 0.002494638
bilingual model 0.002451142
ibm model 0.002384746
segmentation model 0.002354202
word segmentation 0.002314982
chinese word 0.002293321
alignment models 0.0022924679999999998
monolingual model 0.002238436
markov model 0.002232106
process model 0.002212759
foreign word 0.002203304
word probabilities 0.002152522
bigram model 0.0021231040000000002
generative model 0.002119937
empty word 0.002117129
unigram model 0.002112768
unsupervised word 0.0020841080000000003
supervised word 0.002080364
current word 0.0020789890000000003
respective model 0.00207391
pyp model 0.002046683
word boundaries 0.002033403
word seg 0.0020327080000000003
word segmenters 0.002016715
plicit word 0.0020074000000000003
language words 0.001848499
model 0.00177435
machine translation 0.001693399
bilingual models 0.0016605399999999998
accurate alignment 0.001617488
chinese data 0.001562971
language sentence 0.001528649
translation quality 0.001498231
english sentence 0.001463378
ing words 0.001443342
bilingual corpus 0.001410311
hmm models 0.001399551
annotated data 0.001387019
bilingual sentence 0.001380662
conditional probability 0.001374488
bigram models 0.001332502
prior probability 0.001326312
btec data 0.001322505
marginal probability 0.001319827
alignment 0.00130872
glish words 0.0013061370000000002
transition probability 0.001302056
data sets 0.001293186
foreign language 0.001292953
corpora corpus 0.001292749
bilingual method 0.001286078
bilingual segmentation 0.001256644
bilingual corpora 0.0012360219999999998
translation 0.00120359
sentence pair 0.0011863
constraint training 0.001172184
foreign sentence 0.0011720440000000001
language processing 0.0011714149999999999
large corpora 0.001161052
segmentation methods 0.001155039
natural language 0.001153082
empty english 0.001141507
bilingual cor 0.001137712
training set 0.001127395
eign language 0.001104727
test corpus 0.0010982919999999998
method time 0.001096692
previous work 0.001076648
segmentation results 0.0010680009999999998
corpus type 0.001054903
unsegmented corpus 0.0010508689999999998
current sentence 0.001047729
sentence pairs 0.00104536
monolingual segmentation 0.001043938
bilingual uws 0.0010426749999999999
method accuracy 0.001037668
expression corpus 0.001032748
results table 0.001030385
bilingual approaches 0.0010291739999999999
words 0.00102372
monolingual corpora 0.001023316
unsegmented sentence 0.00102122
openmt corpus 0.001020139
training segmenters 0.001019782
bilingual unigram 0.00101521
parameter settings 0.001013014
training resources 0.001011165
bilingual expectation 0.001006225
probability 0.00100621
many languages 0.001005816
bilingual tasks 9.95914E-4
programming method 9.92255E-4
maximum length 9.89841E-4
base distribution 9.88412E-4
segmented sentence 9.86782E-4
models 9.83748E-4
first val 9.80784E-4
parallel corpora 9.78536E-4
method bleu 9.76566E-4
