基于一种文档表示模型的站内搜索引擎设计与实现

摘要

根据全信息理论，认识论信息是语法信息、语义信息和语用信息的三位一体，在信息检索的过程中加入语用信息能有效的提高信息检索的质量。基于查询与内容的文档表示模型较好的利用了语用信息，对站内搜索引擎的查准率的提高有着很好作用；Lucene是一个用java语言开发的开源的全文搜索引擎架构。本文利用Lucene设计和实现一个基于查询与内容的文档表示模型的站内搜索引擎，实验结果表明该模型能有效的提高信息检索的查准率。%According to the comprehensive information theory,epistemology information is the trinity of syntactic information,semantic information and pragmatic information. Making better use of pragmatic information in information retrieval can promote the quality of infor-mation retrieval. A document representation model based on query and content can make better use of pragmatic information,and it is good to promote the precision of the website search engine. Lucene is a open source full text search engine architecture which is devel-oped using java language. We use lucene to design and implement a website engine based on document representation model using query and content. The experiment results show that this model can effectively improve precision rate in information retrieval.