Performance Improvement in Automatic Question Answering System Based on Dependency Term

Authors:
Jianxing Shi;Xiaojie Yuan;Shitao Yu;Hua Ning;Chenying Wang
Affiliations:
Department of Computer Science and Technology, Nankai University, Tianjin, P.R. China 300071;Department of Computer Science and Technology, Nankai University, Tianjin, P.R. China 300071;Department of Computer Science and Technology, Nankai University, Tianjin, P.R. China 300071;Department of Computer Science and Technology, Nankai University, Tianjin, P.R. China 300071;Department of Computer Science and Technology, Nankai University, Tianjin, P.R. China 300071
Venue:
WISM '09 Proceedings of the International Conference on Web Information Systems and Mining
Year:
2009

Citing 5
Cited 0

Simple fast algorithms for the editing distance between trees and related problems

SIAM Journal on Computing
A matching algorithm for measuring the structural similarity between an XML document and a DTD and its applications

Information Systems - Special issue on web data integration
Performance issues and error analysis in an open-domain Question Answering system

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Question answering passage retrieval using dependency relations

Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval
Question Similarity Calculation for FAQ Answering

SKG '07 Proceedings of the Third International Conference on Semantics, Knowledge and Grid

Quantified Score

Hi-index	0.00

Visualization

Abstract

Automatic Question Answering (QA) system has become quite popular in recent years, especially since the QA tracks appeared at Text REtrieval Conference (TREC). However, using only lexical information, the keyword-based information retrieval cannot fully describe the characteristics of natural language, thus the system performance cannot make people satisfied. It is proposed in this paper a definition of dependency term, based on the dependency grammar, employing the natural language dependency structure, as the improvement of the term, to support the typical information retrieval models. It is in fact a solution for a special application in XML information retrieval (XML IR) field. Experiments show that: dependency-term-based information retrieval model effectively describes the characteristics of natural language questions, and improves the performance of automatic question answering system.