tech,22-2-I05-5003,bq |
, WER and PER )
</term>
to building
<term>
|
classifiers
|
</term>
to predict
<term>
semantic equivalence
|
#8359
This paper investigates the utility of applying standard MT evaluation methods (BLEU, NIST, WER and PER) to building classifiers to predict semantic equivalence and entailment. |
other,18-3-I05-5003,bq |
of speech information
</term>
of the
<term>
|
words
|
</term>
contributing to the
<term>
word matches
|
#8385
We 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 the sentence. |
other,12-4-I05-5003,bq |
</term>
are able to produce useful
<term>
|
features
|
</term>
for
<term>
paraphrase classification
|
#8409
Our results show that MT evaluation techniques are able to produce useful features for paraphrase classification and to a lesser extent entailment. |
tech,15-5-I05-5003,bq |
accuracy
</term>
over all of the other
<term>
|
models
|
</term>
used in the experiments . We propose
|
#8435
Our technique gives a substantial improvement in paraphrase classification accuracy over all of the other models used in the experiments. |