Web站点的超链结构挖掘

ＷＷＷ是一个由成千上万个分布在世界各地的Ｗｅｂ站点组成的全球信息系统，每个Ｗｅｂ站点又是一个由许多Ｗｅｂ页构成的信息（子）系统。由于一个文档作者可以通过超链把自己的文档与任意一个已知的Ｗｅｂ页链接起来，而一个 Ｗｅｂ站点上的信息资源又通常是由许多人共同提供的， 因此 Ｗｅｂ站点内的超链链接通常是五花八门、各种各样的，它们可以有各种含义和用途。文章分析了ＷＷＷ系统中超链的使用特征和规律，提出了一个划分超链类型、挖掘站点结构的方法，初步探讨了它在信息收集和查询等方面的应用。

Hyperlink Structure Mining of Web Sites

WWW is a global information system consisting of many Web sites around the world.Each Web site is a subsystem composed of many Web pages.There is no limitation for the writer of a WWW document to set a hyperlink pointing to any other document.Since the documents of a Web sites are often provided by more than one person,the meaning and way of hyperlinks of a Web site is various.In this paper,we analyze the characteristic of the use of hyperlinks in WWW system,present a practical method for recognizing the type of hypertext links and mining the structure of Web sites,and discuss its application on the area of information retrieval.