#7382The task of machine translation (MT) evaluation is closely related to the task ofsentence-level semantic equivalence classification.
measure(ment),12-2-I05-5003,ak
<term>
MT evaluation methods
</term>
(
<term>
BLEU
</term>
,
<term>
NIST
</term>
,
<term>
WER
</term>
#7399This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER and PER) to building classifiers to predict semantic equivalence and entailment.
measure(ment),14-2-I05-5003,ak
methods
</term>
(
<term>
BLEU
</term>
,
<term>
NIST
</term>
,
<term>
WER
</term>
and
<term>
PER
</term>
#7401This paper investigates the utility of applying standard MT evaluation methods (BLEU,NIST, WER and PER) to building classifiers to predict semantic equivalence and entailment.
</term>
and
<term>
PER
</term>
) to building
<term>
#7403This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST,WER and PER) to building classifiers to predict semantic equivalence and entailment.
measure(ment),18-2-I05-5003,ak
<term>
NIST
</term>
,
<term>
WER
</term>
and
<term>
PER
</term>
) to building
<term>
classifiers
</term>
#7405This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER andPER) to building classifiers to predict semantic equivalence and entailment.
tech,22-2-I05-5003,ak
</term>
and
<term>
PER
</term>
) to building
<term>
classifiers
</term>
to predict
<term>
semantic equivalence
#7409This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER and PER) to buildingclassifiers to predict semantic equivalence and entailment.
other,25-2-I05-5003,ak
<term>
classifiers
</term>
to predict
<term>
semantic equivalence
</term>
and
<term>
entailment
</term>
. We also
#7412This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER and PER) to building classifiers to predictsemantic equivalence and entailment.
other,28-2-I05-5003,ak
<term>
semantic equivalence
</term>
and
<term>
entailment
</term>
. We also introduce a novel
<term>
#7415This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER and PER) to building classifiers to predict semantic equivalence andentailment.
tech,5-3-I05-5003,ak
</term>
. We also introduce a novel
<term>
classification method
</term>
based on
<term>
PER
</term>
which leverages
#7422We also introduce a novelclassification method based on PER which leverages part of speech information of the words contributing to the word matches and non-matches in the sentence.
measure(ment),9-3-I05-5003,ak
classification method
</term>
based on
<term>
PER
</term>
which leverages
<term>
part of speech
#7426We also introduce a novel classification method based onPER which leverages part of speech information of the words contributing to the word matches and non-matches in the sentence.
other,12-3-I05-5003,ak
on
<term>
PER
</term>
which leverages
<term>
part of speech information
</term>
of the words contributing to the
<term>
#7429We also introduce a novel classification method based on PER which leveragespart of speech information of the words contributing to the word matches and non-matches in the sentence.
other,22-3-I05-5003,ak
</term>
of the words contributing to the
<term>
word matches
</term>
and non-matches in the
<term>
sentence
#7439We also introduce a novel classification method based on PER which leverages part of speech information of the words contributing to theword matches and non-matches in the sentence.
other,28-3-I05-5003,ak
matches
</term>
and non-matches in the
<term>
sentence
</term>
. Our results show that
<term>
MT evaluation
#7445We also introduce a novel classification method based on PER which leverages part of speech information of the words contributing to the word matches and non-matches in thesentence.
tech,14-4-I05-5003,ak
able to produce useful features for
<term>
paraphrase classification
</term>
and to a lesser extent
<term>
entailment
#7461Our results show that MT evaluation techniques are able to produce useful features forparaphrase classification and to a lesser extent entailment.
other,21-4-I05-5003,ak
classification
</term>
and to a lesser extent
<term>
entailment
</term>
. Our technique gives a substantial
#7468Our results show that MT evaluation techniques are able to produce useful features for paraphrase classification and to a lesser extententailment.
tech,7-5-I05-5003,ak
gives a substantial improvement in
<term>
paraphrase classification accuracy
</term>
over all of the other models used
#7477Our technique gives a substantial improvement inparaphrase classification accuracy over all of the other models used in the experiments.