MBRTE: Modified BLEU Algorithm for Recognition Textual Entailment

DOI

10.2991/aisr.k.200424.066How to use a DOI?

Keywords

MBRTE, BLEU, textual entailment

Abstract

The BLEU algorithm has become a standard evaluation tool for Machine Translation (e.g Goole Translate). The advantages of this algorithm are simple and fast. This algorithm is quite widely applied in fields other than MT, including in the Recognition Textual Entailment (RTE). The implementation of BLEU algorithm to the RTE requires some modification because of several problems. One of these problems is related to the sensitivity of BLEU scores. This research has modified BLEU to overcome these problems. Experimental data set contains 200 hypotheses with each hypothesis have some text. The experimental results obtained an accuracy of 83,82 % for modified BLEU. While the accuracy results using BLEU without modification is obtained at 77,50%. So that the accuracy increases of 6,32%.

Copyright

© 2020, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Abdiansah ABDIANSAH
AU  - Alvi Syahrini UTAMI
PY  - 2020
DA  - 2020/05/06
TI  - MBRTE: Modified BLEU Algorithm for Recognition Textual Entailment
BT  - Proceedings of the Sriwijaya International Conference on Information Technology and Its Applications (SICONIAN 2019)
PB  - Atlantis Press
SP  - 432
EP  - 436
SN  - 1951-6851
UR  - https://doi.org/10.2991/aisr.k.200424.066
DO  - 10.2991/aisr.k.200424.066
ID  - ABDIANSAH2020
ER  -