metrics metrics 0.00339612
evaluation metric 0.00333288
human evaluation 0.0033219599999999997
evaluation metrics 0.00320048
automatic metrics 0.002399616
only metric 0.002358538
tion metric 0.00235196
word order 0.002333664
gtm metric 0.002293937
human judges 0.002285599
human judgements 0.002255936
source word 0.002255701
word error 0.002246525
language model 0.002239145
nist metric 0.002212066
parsing metric 0.002204119
automatic evaluation 0.002203976
evaluation system 0.002200961
human judge 0.002187817
syntactic metrics 0.0021638539999999998
human evaluations 0.002140617
wer word 0.002089971
sentence correlation 0.00208373
tomatic metrics 0.00205945
uation metrics 0.002039421
machine translation 0.002014357
translation error 0.001942955
evaluation experiments 0.001928669
dependency evaluation 0.00191552
ter translation 0.0018815869999999999
tomatic evaluation 0.00186381
metric 0.00183046
data sentence 0.001827452
sentence level 0.001809007
language models 0.001805079
dard evaluation 0.001802586
evaluation software 0.001800364
evaluation procedure 0.001800364
translation field 0.0017846059999999998
sentence corpus 0.001777223
corpus correlation 0.001776773
metrics 0.00169806
english data 0.0016675169999999999
test corpus 0.0015923740000000001
rank correlation 0.001565117
average sentence 0.001564944
sentence length 0.001543455
evaluation 0.00150242
corpus level 0.00150205
translation 0.00148005
individual sentence 0.001475022
correlation coefficient 0.00147501
same data 0.001458576
same level 0.0014401309999999999
linguistic level 0.0014336689999999998
test set 0.001424221
correlation coefficients 0.0013885019999999998
pearson correlation 0.001373828
grammatical sentences 0.001356639
significant correlation 0.0013409399999999999
english output 0.0013277900000000001
language 0.00132687
ranking model 0.001317395
average score 0.0013155369999999999
different tasks 0.001278648
plicate words 0.001269839
lexical resource 0.001263218
guage model 0.0012365940000000001
english summarisation 0.001235544
english weather 0.001231804
same order 0.001223258
first task 0.0012099839999999999
corpus string 0.001207154
english paraphrases 0.00120486
original corpus 0.001197695
german corpus 0.001176149
ranking task 0.001172941
standard set 0.0011658670000000001
dependency parser 0.0011503
system output 0.001144176
other text 0.0011413299999999999
judgement task 0.001135601
corpus corr 0.00113458
ing system 0.001119244
absolute score 0.001116321
tree representation 0.0011091550000000001
good measures 0.0011043749999999999
ranking system 0.0011036610000000001
level corre 0.001094339
naturalness task 0.001090969
average scores 0.001079749
other languages 0.001069624
realisation system 0.001058958
order ones 0.0010571259999999998
automatic evalua 0.0010555270000000001
sentence 0.00104209
correlation 0.00104164
lfg parser 0.0010366870000000001
automatic eval 0.0010268389999999999
automatic evaluations 0.001022633
