transliteration model 0.00457965
learning transliteration 0.00434922
chinese transliteration 0.003474694
transliteration pairs 0.003445822
transliteration probability 0.003408078
new transliteration 0.0033629999999999997
phonetic transliteration 0.003332375
transliteration candidates 0.003259732
transliteration pair 0.003186966
transliteration lexicon 0.003178972
machine transliteration 0.003178223
candidate transliteration 0.0031684969999999997
transliteration modeling 0.0031521889999999997
transliteration errors 0.003109128
qualified transliteration 0.003099505
automatic transliteration 0.003073849
date transliteration 0.003063724
regular transliteration 0.003056676
transliteration criteria 0.003054593
transliteration constituent 0.003044247
transliteration entry 0.003028823
transliteration lexicons 0.003028082
transliteration fall 0.0030213379999999997
model english 0.0029407900000000004
language model 0.0027667660000000004
transliteration 0.0027459
psm model 0.00255356
translation model 0.002506675
english word 0.00238038
learning approach 0.002338437
psm learning 0.00232313
model accuracy 0.002316326
learning algorithm 0.002268411
probabilistic model 0.002213868
active learning 0.0021985520000000003
similarity model 0.002193679
generative model 0.002188372
unsupervised learning 0.0021754920000000002
model let 0.002155653
channel model 0.002152109
learning process 0.002127068
learning framework 0.0020726900000000003
supervised learning 0.002069275
learning lexicon 0.002036392
lexicon learning 0.002036392
machine learning 0.002035643
training data 0.002024561
chinese word 0.002002134
learning strategy 0.001953247
effective learning 0.0019298920000000001
adaptive learning 0.0019292620000000002
learning strategies 0.001926007
incremental learning 0.001877947
learning progress 0.001877947
model 0.00183375
english words 0.001823954
data labeling 0.0017760509999999998
new english 0.00172414
single word 0.001657073
foreign word 0.001650542
labeled data 0.0016075969999999999
learning 0.00160332
word frequencies 0.001567305
noisy data 0.001563948
unlabelled data 0.001558732
translation problem 0.001520872
original english 0.0015197370000000002
different approach 0.00150255
english syllables 0.001495016
english sound 0.0014638560000000001
training samples 0.001441384
source language 0.0014345970000000001
chinese character 0.001418904
english graphemes 0.001385116
english appositives 0.001383096
translation pairs 0.001372847
original language 0.0013457130000000001
chinese transliterations 0.001330355
confidence candidates 0.001319754
high confidence 0.001318794
new pairs 0.001317022
mapping approach 0.0012647589999999998
high performance 0.001258145
information source 0.001247381
selection method 0.0012459020000000001
character string 0.0012415809999999998
approach models 0.00123285
confidence score 0.001231499
native language 0.00121794
alphabetical language 0.001208263
syllabic language 0.001208263
bilingual words 0.001202525
labeling samples 0.001192874
live corpus 0.001192743
training lexicon 0.001186003
confidence scores 0.001172972
similar results 0.001147365
new knowledge 0.001139798
query results 0.001134615
proper names 0.001133659
