Using Machine Translation for Recognizing Textual Entailment in Vietnamese Language

Abstract

Recognizing Textual Entailment (RTE) is a fundamental task in Natural Language Understanding. The task is to determine whether the meaning of a text can be inferred from the meaning of the other one. This paper explores the use of Machine Translation (MT) in recognizing textual entailment in texts written in Vietnamese. We present two methods of using Machine Translation for Vietnamese RTE. The first method integrates a MT component into front-end of an English RTE system. The second method uses a MT component to produce English translation of Vietnamese RTE data, and both original Vietnamese data and its translation are used to learn an entailment classifier. Experimental results achieve on Vietnamese RTE corpus built from RTE3 data set suggest that Machine Translation can help to improve Vietnamese RTE.