language model 0.00298485
sms word 0.00288356
sms language 0.00257518
translation model 0.00247688
normalization model 0.0023713189999999998
english word 0.002359702
sms words 0.002158111
model training 0.0021525809999999998
same word 0.002147328
word set 0.002073101
sms translation 0.00206721
english language 0.002051322
mapping model 0.001994829
statistical model 0.001976958
sms normalization 0.0019616489999999998
specific word 0.001935927
tion model 0.001916007
above model 0.001874489
channel model 0.0018677029999999999
english sms 0.001865322
malization model 0.001844996
tical model 0.0018367029999999999
group model 0.0018356709999999999
language modeling 0.001768939
sms text 0.001721574
bigram language 0.0016815859999999999
sms sentence 0.001657492
sms corpus 0.0016557759999999999
english words 0.001634253
srilm language 0.001619831
parallel sms 0.00161681
lish language 0.001615196
language expres 0.001612077
body language 0.001612077
saging language 0.001612077
model 0.00160426
sms domain 0.001567377
sms texts 0.0015592549999999998
chinese sms 0.001535624
simple sms 0.001528619
variation sms 0.001518088
sms chat 0.001516099
translation problem 0.001502394
sms message 0.001501676
sms messages 0.001489046
sms transla 0.001480751
sms lingo 0.001477506
sms normaliza 0.001473912
single sms 0.0014724579999999998
sms application 0.0014704849999999999
sms users 0.001462408
words alignment 0.00144148
raw sms 0.0014381539999999999
sms our 0.001438092
young sms 0.001425949
sms lingoes 0.001425949
separate sms 0.001425949
mercial sms 0.001425949
normalization problem 0.001396833
language 0.00138059
standard words 0.001371439
translation system 0.001365098
many words 0.0013227669999999999
machine translation 0.00130159
text normalization 0.001294043
case words 0.001283794
translation output 0.001278218
normalization results 0.0012725549999999999
prior distribution 0.001271721
training data 0.0012622710000000001
oov words 0.0012529009999999998
translation bleu 0.001243014
such systems 0.00122703
translation performance 0.001222635
unseen words 0.001204243
lish words 0.001198127
english text 0.001197716
erroneous words 0.001196809
normalization pairs 0.001177195
translation technique 0.0011540259999999998
normalization task 0.001140472
such approaches 0.001138504
parallel data 0.00113617
english sentence 0.001133634
posteriori distribution 0.001129485
translation accuracy 0.0011292519999999999
translation quality 0.0011275439999999999
translation sys 0.001120909
translation applications 0.0011191819999999998
translation prob 0.001112587
phrase alignment 0.001109854
consensus translation 0.001104583
data set 0.0010980809999999999
phrase segmentation 0.00109736
normalization pair 0.0010848149999999998
target text 0.001075631
normalization candidate 0.001073875
such irregulari 0.001073254
such irregularities 0.001073254
mapping probability 0.001060194
