transliteration model 0.00470925
name transliteration 0.004061634
transliteration data 0.004017440000000001
transliteration results 0.003839144
transliteration models 0.0037485730000000003
transliteration candidates 0.003665391
transliteration approaches 0.003623545
transliteration modeling 0.003608909
discriminative transliteration 0.003591636
transliteration accuracy 0.00358165
previous transliteration 0.003561443
global transliteration 0.0035527700000000002
language model 0.00353939
single transliteration 0.003534764
local transliteration 0.003529865
joint transliteration 0.003503927
testing transliteration 0.003474075
transliteration dictionary 0.003462134
korean transliteration 0.003458131
transliteration dataset 0.003438404
transliteration application 0.003434955
transliteration experiments 0.003426446
transliteration prediction 0.0034098180000000002
transliteration speed 0.003394191
transliteration 0.00310003
language pairs 0.002624494
language modeling 0.002439049
foreign language 0.00234504
natural language 0.002267799
english names 0.002078687
probabilistic model 0.001971774
language 0.00193017
eration model 0.001917129
model applica 0.001899415
different languages 0.001831137
correct english 0.001744546
name pairs 0.001655928
english candidate 0.001633874
feature representation 0.001626193
model 0.00160922
modeling feature 0.0015873989999999998
feature space 0.001576456
english spelling 0.0015710239999999999
global feature 0.00153126
local feature 0.0015083549999999999
ing feature 0.0015023699999999998
english translitera 0.001476029
feature spaces 0.001450503
legitimate english 0.001438104
english speakers 0.001438068
string pairs 0.001419221
global features 0.001416932
feature ϕlen 0.0014002679999999999
local features 0.001394027
difference feature 0.0013920599999999999
foreign name 0.001376474
feature φlen 0.001375289
pertinent feature 0.001368844
same training 0.001368474
binary features 0.001362153
original name 0.001359897
ing names 0.001355787
foreign names 0.001346807
arabic training 0.0013449159999999998
different lan 0.001286324
different values 0.001267659
modeling approach 0.001262501
extraneous features 0.001255623
learning problem 0.0012380870000000001
tween names 0.001222284
probability models 0.001220257
data the 0.001218759
machine learning 0.0012125740000000001
information extraction 0.0012071389999999999
experimental results 0.001158124
training examples 0.001155697
english 0.00114675
training sample 0.001138712
other hand 0.001126864
perceptron training 0.001122928
global set 0.001116372
probabilistic approach 0.001116176
the training 0.001079558
feature 0.00107852
local models 0.001078378
relative distance 0.001068401
criminative approach 0.001049515
approach achiev 0.00104506
superior performance 0.001042664
korean words 0.0010412680000000001
empty string 0.001040835
literation performance 0.001034861
modeling approaches 0.0010323939999999998
binary learning 0.00102762
decoding process 0.0010191760000000001
string termination 0.001017919
important task 0.001017226
learning problems 0.0010165
count information 0.001015527
discriminative approaches 0.001015121
