evaluation metrics 0.00372906
evaluation metric 0.00355878
translation evaluation 0.0033272600000000003
human translations 0.003054865
automatic metrics 0.002971043
different metrics 0.0029301730000000003
human reference 0.002776792
other metrics 0.002737418
automatic evaluation 0.002634723
human judgments 0.00262755
human judges 0.002586784
human ref 0.002557013
standard metrics 0.002555266
human assessment 0.002553126
regression metrics 0.002493766
individual metrics 0.0024836270000000004
overall human 0.002476765
human assessments 0.002473363
evaluation task 0.002468663
human likeness 0.00243563
human judg 0.002423719
trained metrics 0.0024137250000000002
metrics gen 0.002408879
evaluation score 0.0024031169999999998
reliable metrics 0.0024007800000000004
level evaluation 0.002397945
ual metrics 0.002384539
corresponding metrics 0.0023797410000000003
reference translation 0.002345012
translation sentences 0.0023312000000000003
regression metric 0.002323486
chain metric 0.002293079
translation output 0.002272779
evaluation judgments 0.0022612500000000002
translation chinese 0.002248149
translation quality 0.002241457
composite metric 0.0022395970000000003
rived metric 0.002208991
ite metric 0.002208991
metric design 0.002208991
evaluation reliability 0.0021693
evaluation question 0.002165757
machine translation 0.002162391
translation outputs 0.002139579
tomatic evaluation 0.002111447
evaluation problem 0.002094688
evaluation correla 0.002076011
human 0.00206267
matic evaluation 0.002050389
efficient evaluation 0.002046623
evaluation methodologies 0.002040953
metrics 0.00203269
multiple translation 0.002024089
translation sound 0.00199185
sentential translation 0.001975962
translation qualities 0.001975962
translation quali 0.001975962
metric 0.00186241
test data 0.0018343460000000002
different model 0.001818495
test systems 0.001719603
same data 0.00171615
reference translations 0.001706317
different source 0.001699748
evaluation 0.00169637
common word 0.0016936120000000002
training data 0.001685636
linguistic data 0.001671481
test sentences 0.001671126
same source 0.001654885
test set 0.0016482369999999999
source language 0.001639286
translation 0.00163089
feature model 0.001617236
model features 0.001614645
different correlations 0.001571932
word class 0.0015535520000000001
machine translations 0.001523696
training sentences 0.001522416
word error 0.001506598
source sentences 0.001502575
training set 0.001499527
linear correlation 0.001487712
sentence level 0.0014815
chinese data 0.0014807890000000002
good system 0.001453677
same accuracy 0.001394026
regression model 0.001382088
feature set 0.001373645
classification training 0.0013721599999999999
system development 0.001370233
single system 0.001364764
erence translations 0.001358232
corpus features 0.001354358
assessment data 0.0013539860000000002
rank correlation 0.00134981
likeness test 0.001343776
different svm 0.001337868
test instance 0.001331414
source text 0.0013297909999999999
