XML关键字检索中推断用户需求信息对象的方法XObject

XObject: An XML Keyword Search Method Based on Structural Retrieval

基于关键字查询的XML检索技术,因为它的用户友好性,近几年得到了信息检索领域研究者的广泛关注。但是由于关键字缺少XML结构语义信息,检索结果和用户需求存在较大偏差。而基于结构的XML检索,用户不仅需要了解XML内部结构,还要掌握结构查询语言,导致用户难以提出准确描述查询意图的查询表达式。文章结合二者研究方向之长,提出一种基于关键字的结构查询方法XObject。XObject用面向对象的思想,分析查询关键字和XML的结构信息,推断用户查询的信息对象,构建一组结构查询语句,再通过现有的结构检索系统,实现查询。经在多个XML实际数据集上验证,结果表明,XObject方法具有很高的查全率和查准率,尤其是和经典的关键字查询方法SLCA相比,XObject方法查准率有明显提高。

Aim.The introduction of the full paper points out the two reasons why the user can not construct effective structural queries;as a result,the queries over XML data often get useless answers.To address this challenge,this paper proposes a new approach for automatically constructing XML structural queries,named XObject.Section 1 reviews past literature.Section 3 explains the XObject method;its core consists of：（1） the generation of a structural query from a keywords query by analyzing the keywords and the schema of XML data;（2） the evaluation of the structural query on XML data with existing XQuery or XPath search engine.Section 4 presents our experimental work;the experimental results for XObject and SLCA,presented in Figs.3 and 4,and their analysis show preliminarily that,compared with SLCA method,XObject method is a little better in recall rate but is much better in precision rate.