基于相关术语集的搜索引擎选择

1 引言 Web从1991年出现以来,已经发展成为一个巨大的全球化信息空间,而且其信息容量仍在以指数形式飞速增长。面对海量Web信息资源,如何有效地检索Web信息,以帮助用户从大量文档信息集合中找到对给定查询请求有用的文档子集,也就成为一项重要而迫切的研究课题。

Search Engines Selection Based on Relevance Terms

Metasearch can effectively search distributed immense electronic resources. It is built on top of several search engines, providing user with uniform access to these engines. Metasearch first passes user's query to underlying useful search engines, and then collects and reorganizes the results from the search engines used. It is called search engines selection when metasearch selects underlying useful search engines. In this paper, we present a statistical method based on relevance terms to estimate the usefulness of a search engine for any given query, which is suitable for both Boolean query and vector query. Experimental results indicate that the proposed estimation method is quite accurate, especially when the critical similarity is high between the query and the results.