evaluation measures 0.00256155
evaluation measure 0.00239725
other measures 0.002133678
similarity measures 0.002024224
evaluation results 0.0019244550000000003
human references 0.001914764
human translations 0.0018800660000000001
several measures 0.001873785
human reference 0.001863813
similarity measure 0.001859924
lexical measures 0.001857709
measure set 0.001852088
linguistic measures 0.001846345
single measures 0.001840743
automatic evaluation 0.0018252160000000002
text evaluation 0.001811475
evaluation similarity 0.0018109740000000001
new measures 0.001789654
human judgements 0.001758705
tion measures 0.0017578350000000001
standard measures 0.001754003
individual measures 0.001734765
common measure 0.001734508
additional measures 0.0017318799999999999
reliability measure 0.001730271
human assessors 0.001712333
human judges 0.001709893
measures bleu 0.001709635
human assessments 0.001709262
human likeness 0.001696443
original measures 0.00169624
human acceptability 0.0016943330000000001
lexical measure 0.0016934089999999999
evaluation reliability 0.0016813210000000002
novel measures 0.00168081
single measure 0.001676443
rouge measures 0.001675911
eral measures 0.0016559090000000001
sophisticated measures 0.001654702
combined measures 0.001653185
heterogeneous measures 0.001650162
tional measures 0.00164347
erogeneous measures 0.001639771
comparable measures 0.001636398
neous measures 0.001635111
cludes measures 0.001635111
new measure 0.0016253539999999999
translation correlation 0.0016074470000000001
tion measure 0.001593535
possible measure 0.001589917
measure sets 0.001578815
measure combination 0.0015723059999999999
individual measure 0.001570465
additional measure 0.0015675799999999998
measure pairs 0.001566729
measure bleu 0.001545335
basic measure 0.001543209
original measure 0.0015319399999999999
measure meteor 0.001520638
rouge measure 0.001511611
evaluation result 0.0015031940000000002
summarization evaluation 0.0014928460000000001
shared evaluation 0.0014873010000000001
heterogeneous measure 0.001485862
measure combi 0.001485827
perfect measure 0.001484055
measure satisfies 0.001482778
measure nist 0.001480269
reliable measure 0.001474862
current evaluation 0.001474707
global measure 0.0014745
evaluation tasks 0.001474285
liable measure 0.001474003
measure satis 0.001470859
dividual measure 0.001470859
evaluation process 0.001467806
other words 0.001435088
unanimous evaluation 0.001426821
evaluation campaigns 0.0014253760000000001
corroborating evaluation 0.001424941
ditional evaluation 0.0014226330000000002
duc evaluation 0.0014217540000000001
such results 0.001401351
system translation 0.0013900560000000002
automatic translation 0.001387813
measures 0.0013874
different values 0.001277886
different quality 0.001265434
similar results 0.001241976
machine translation 0.001229441
measure 0.0012231
high correlation 0.0012165729999999999
different document 0.001211817
empirical results 0.0011996160000000001
spearman correlation 0.0011958099999999998
evaluation 0.00117415
low correlation 0.001168016
system level 0.0011441329999999999
pearson correlation 0.001133867
other metrics 0.00112248
