摘要
网络结构挖掘是以超链接分析为基础,从链接结构中获取有用的知识,利用这些知识,重新组织结构,使内容逻辑结构更加合理。深入研究现有的网络结构挖掘系统,并在对其核心算法PageRank和HITS中所存在的问题作了详细分析的基础上提出了自己的改进算法,主要是对每个网页定义这三个参数:PageRank,Authority,Hub,并进行分析与优化,以便得到更好的查询结果,最后设计了一个改进网络结构挖掘系统原型,根据实验结果进行分析。
Web structure mining is based on hyperlink analysis. It has been gained useful information from man - made links structure. Pages can be sorted making use of it. And important content pages can also be found so that can reform web structure to gain better contant structure. And go deep into researching algorithm used in existing web structure system. And improves its core algorithm. Mainly analyze and optimize three data of page： PageRank and Authority and Hub so that can gain the best query result. Also design an improvement web structure system prototype with experimental result and data analysis at last.
出处
《计算机技术与发展》
2009年第5期41-44,共4页
Computer Technology and Development
基金
安徽省自然科学研究项目(KJ2007B245)
