Abstract: Semistructured data based on Web is taken more and more attention now. This paper analyzes data retrieval of XML currently, and proposes a kind of path index technology which have been combined with inverted file to implement retrieval both on context and structure. This method needs scan only once to file base, which can reduce user query time with a large number.

Key words: XML, Information retrieval, Index, Inverted file

摘要： 当前基于Web的半结构化数据越来越受到重视。该文分析了当前对XML数据检索的相关工作，提出了一种路径索引技术，并将之无缝结合了基于文本的倒排索引文档，以实现XML文档的内容和结构的双重检索。该方法只需要对文档库扫描一次，可以大幅度降低用户查询时间。

关键词: XML, 信息检索, 索引, 倒排文档

CLC Number: 

• TP312