网页链接繁殖在搜索引擎资源发现中的应用  

Application of hyperlinks reproduction in meta search resource discovery

作　　者：陈爽[1] 钱榕[2] 陈福[2] 李素[3] 

机构地区：[1]西北工业大学计算机学院,西安710072 [2]北京科技大学信息工程学院,北京100083 [3]北京工商大学计算机学院,北京100037

出　　处：《计算机工程与应用》2007年第36期177-180,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.60673160);北京市自然科学基金(the NaturalScience Foundation of Beijing of China under Grant No.402200)

摘　　要：为解决搜索引擎返回结果数量上的限制,扩展了元搜索技术,提出链接群落、链接繁殖的概念,并与生物群落进行了对比。链接繁殖的思想是首先将多个搜索引擎返回的结果作为起始信息源,利用预定义的繁殖规则,优化并整合搜索结果,对链接所指网页的链接进行分析,繁殖出更多的相关信息源。在分析不同的搜索引擎结果集时,系统根据不同搜索引擎直接与繁殖发现信息源的能力与质量,动态调整繁殖的链接的优先次序。经过实验验证,链接繁殖可以大大扩展通过搜索引擎发现主题信息源的数量。In order to solve the problem of quantity restriction of the search engine results,this paper presents a concept of link community by expanding Meta-Search Engine technique and compares with the biological community.The idea of link reproduction is taking many search engines results as the initial source information at first,then optimizing and integrating the search results by using predefined reproduction rules,and then reproducing more related information sources by analyzing the links of the WebPages.When analyzing result collections from different search engine,the system directly reproduces and discovers the ability and the quality of information source according to the different search engine,and dynamically adjusts the precedence of reproduced links.The experiments prove that the link reproduction may greatly increase the quantity of information sources when discovering subjects through the search engine.

关 键 词：主题发现 元搜索 链接分析 链接繁殖 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]