linguistic features 0.002712982
translation system 0.0024928800000000003
human translation 0.002434746
language model 0.00237389
linguistic feature 0.002317172
perplexity features 0.00225492
features perplexity 0.00225492
oov features 0.00212815
abstract features 0.00208171
miscellaneous features 0.002076112
density features 0.002075964
annotated features 0.0020735090000000003
features yields 0.002072949
machine translation 0.002071618
discriminatory features 0.0020699720000000002
system evaluation 0.00206917
human evaluation 0.002011036
reference translation 0.0019429500000000001
training sentences 0.0019186960000000001
language words 0.001889964
input feature 0.001836157
features 0.00180115
english sentences 0.001787092
test sentences 0.0017763800000000001
target language 0.001753585
source language 0.001739984
training data 0.0017365129999999999
candidate feature 0.0017285669999999999
language models 0.00171507
function word 0.001709609
output sentences 0.001643892
reference sentences 0.0016410399999999999
english sentence 0.001636775
language generation 0.001609834
problem sentences 0.001601053
classification system 0.001599495
test data 0.0015941969999999999
human translations 0.001568855
system output 0.001567802
automatic evaluation 0.0015676919999999999
coarse model 0.00155953
output quality 0.001536502
statistical language 0.001529895
individual sentences 0.0015197449999999999
human reference 0.0015068159999999998
word density 0.001505134
system development 0.001482969
second language 0.001477404
linguistic information 0.001439275
translation 0.00143544
sentence length 0.001432114
system overall 0.001420153
human post 0.001407425
feature 0.00140534
spanish sentences 0.001404854
natural language 0.0013942520000000001
language processing 0.001388673
sentence pairs 0.0013860749999999999
language modeling 0.001383816
building language 0.0013726
processing system 0.001362613
manual evaluation 0.001344918
automated evaluation 0.00132264
data table 0.00132259
full evaluation 0.001319798
sentence basis 0.00129492
evaluation methodology 0.001294314
model 0.00129039
function words 0.001285753
data point 0.001282714
human ability 0.001274796
sentence templates 0.0012621009999999998
sentence template 0.0012540149999999998
same set 0.0012529070000000002
data our 0.001238509
special case 0.001235911
linguistic motivation 0.001228559
machine translations 0.0012057270000000002
content words 0.001176413
machine learning 0.0011699800000000002
english output 0.001163924
english reference 0.001161072
training cases 0.001144344
average accuracy 0.001141885
sentences 0.00113353
oov words 0.001133464
perplexity score 0.001123853
syntactic parse 0.001106905
training texts 0.001100907
additional training 0.00110027
training classifiers 0.0010875960000000001
language 0.0010835
reference translations 0.001077059
classification task 0.001075969
learning classification 0.001075857
syntactic template 0.00107304
baseline accuracy 0.00106651
system 0.00105744
goodness metrics 0.001035112
quality 0.00102614
