training data 0.00329355
data selection 0.00305844
different data 0.002698507
translation performance 0.00222987
data resources 0.002210339
above data 0.002192752
parallel training 0.0021682159999999997
available data 0.002155002
training set 0.00215455
smt training 0.002150324
training corpus 0.0021301889999999998
translation probability 0.002119903
translation task 0.002119654
data configuration 0.002105897
data configurations 0.002102576
translation error 0.001987698
translation quality 0.001984732
potential translation 0.001963173
possible training 0.001962968
selection algorithm 0.001957586
machine translation 0.001949499
translation resources 0.001937859
manual translation 0.001894906
translation hypothesis 0.0018947789999999999
posterior translation 0.0018907919999999999
translation tasks 0.001885962
translation edit 0.001849082
translation errors 0.0018442859999999999
uate translation 0.0018325009999999998
translation difficulty 0.001830651
candidate sentence 0.001784715
pool training 0.0017777959999999999
training corpora 0.0017712749999999999
source sentences 0.001763358
comparator training 0.001753738
pairwise sentence 0.0017517540000000001
sentence pairs 0.001741795
seed training 0.0017389769999999998
labeled training 0.001733928
selection approach 0.001712421
feature function 0.0016893479999999998
other selection 0.0016853389999999999
training partition 0.0016763919999999999
training entries 0.00167415
sample selection 0.0016691969999999999
random selection 0.00161149
choice sentence 0.001603471
translation 0.00158783
sentence compari 0.0015864030000000001
active selection 0.00157026
candidate sentences 0.001551375
selection strategy 0.0015491799999999998
greedy selection 0.0015387929999999999
above selection 0.001530572
test set 0.001524108
selection technique 0.001508844
discriminative selection 0.001508611
selection methods 0.001505916
source language 0.001493629
unsupervised selection 0.0014827669999999999
selection iterations 0.001473556
selection techniques 0.001470476
bleu score 0.001464226
overall selection 0.001462638
incremental selection 0.0014565799999999999
selection strategies 0.0014508379999999999
dissimilarity selection 0.001449868
ple selection 0.00144838
selection tech 0.001447011
selection criteria 0.00144251
passive selection 0.001439999
training 0.00143324
parallel corpus 0.0014319250000000001
source words 0.001423223
chosen sentences 0.0014072989999999999
constituent sentences 0.001393746
subsequent sentences 0.001386794
entropy model 0.001384438
identical sentences 0.001383191
comparator function 0.001367758
quent sentences 0.001365711
select sentences 0.001361241
smt performance 0.001359124
pick sentences 0.00135671
source corpus 0.0013487170000000001
sentence 0.00134493
feature functions 0.001316167
ture function 0.0012989219999999999
human translations 0.001282038
ranking model 0.001253998
multiple language 0.001245204
ing corpus 0.0012424889999999998
language pairs 0.0012387259999999999
bleu scores 0.001233662
smt system 0.001231736
word align 0.001211656
other smt 0.001204293
feature vector 0.001204181
selection 0.00119813
corpus size 0.001179652
