角色反演算法在问答系统中的应用  

Application of Role Inverse Algorithm in Question Answering System

作　　者：党琰[1] 张冬茉[1] 李芳[1] 

机构地区：[1]上海交通大学计算机科学与工程系,上海200030

出　　处：《计算机工程与应用》2004年第36期213-217,共5页Computer Engineering and Applications

基　　金：国家自然科学基金资助(编号:60083003)

摘　　要：该文介绍了如何将角色反演算法的思想用在多信息源多语种问答系统中来构建句法分析器。常用的句法分析算法由于受到语法规模大小的限制,一般都不能有效地应用到实际的自然语言处理当中。角色反演算法思想是将Chart算法的高空间效率和广义LR算法的高时间效率有效地结合起来,从而大大提高了综合的分析效率。基于多信息源多语种的问答系统,拥有大规模语法(上万条语法规则),通过引入角色反演算法思想,可以分别在问句分析模块和答句生成模块中有效地完成问句和文本答案候选文档的句法分析。This paper introduces how to use the concept of Role Inverse Algorithm in order to build syntactic parser in the various information resources based multi-lingual question answering system.The most popular parsing algorithms can seldom be effectively used in natural language processing because of the large scale of the grammars.The idea of the Role Inverse Algorithm is to combine the high space efficiency of Chart parsing algorithm and the high time efficiency of GLR parsing algorithm,so the comprehensive parsing efficiency can be highly improved.The various information resources based multi-lingual question answering system contains large scale grammars,up to thousands upon thousands rules.With the introduction of the Role Inverse Algorithm conception,people can accomplish the syntactic parsing of both questions and candidate answers in question analysis module and answer creation module in the system effectively.

关 键 词：角色反演算法 句法分析 多信息源 多语种 问答系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]