Implementation Method of Answering Engine for Vietnamese Questions in Reading Answering System Model (RASM)

Abstract

This paper aims to present our method of implementation which is introduced to build the answering engine of RASM (Reading Answering System Model). RASM is built to run on Prolog's engine and can read many forms of Vietnamese news titles of ICTNEWS (http://ictnews.vn) to answer several types of Vietnamese questions. In RASM, the answering engine is the central kernel. The RASM's answering engine includes two principal elements: i) first element is a computational grammar which is defined in DCG (Definite Clause Grammar) to determine the semantic representation of Vietnamese questions, ii) second element is a seeking method to answer the questions.