【作者】 庄玲盈；

【导师】 杨丹；

【作者基本信息】 重庆大学 ， 计算机软件与理论， 2006， 硕士

【摘要】 随着Internet的不断发展,用户对Internet提供信息的主体——Web站点的要求越来越高。这种要求不但体现在站点信息的广度和深度上,更体现在网站的内容和结构是否随用户的兴趣而针对性地改变,这就引出了“个性化”这个研究课题。Web数据的挖掘以及在此基础上建立的自适应网站对满足用户的个性化要求有很好的应用效果。个性化自适应网站的适应性对用户透明,它不同于用户定制的网站,不需要用户额外的输入而自动的获取用户的兴趣爱好,从而提供个性化的服务。基于自适应网站的上述优点,在电子商务日臻成熟和越来越看重以人为本的今天,它有着广泛的应用前景。本文研究的目的是Web日志挖掘在自适应网站中实现动态链接自动生成的功能。论文首先介绍了当前数据挖掘领域的发展现状;其次讨论了Web数据挖掘技术,包括Web数据挖掘的分类及各类Web数据挖掘的实现技术;接着分析研究了Web日志挖掘,包括对Web日志挖掘过程模型以及相关算法的分析;然后论述对自适应网站系统的设计以及自动增加动态链接功能的研究实现,本文采用Cookie技术和扩充日志属性方法进行用户识别,采用聚类算法对用户访问模式进行挖掘,并通过在网页中嵌入Asp代码的方式,用程序实现自动在页面中增加动态链接;最后对本文进行了总结,对Web挖掘和自适应网站的未来进行了展望。更多还原

【Abstract】 With the continuous development of the Internet, the user’s requirements are becoming higher and higher. The requirement is not only reflected in the breadth and depth of information, but also more incarnated in the self-adaptative Website. That leads to the subject,“personalization”. What’s more, the“Self-adaptive Website”(which will be referred to SAWS), which is based on the technology of“Web Mining”, plays a very good effect in the user’s requirements. Meanwhile, SAWS is transparent to the user, which is differ from the customized Website, it does not need user’s additional input but automatic accesses users’ interest. Today, with the more and more mature e-commerce, the SAWS, which has advantages as above, will has broad application prospects.The purpose of the paper is realizing the automaticly dynamic hyperlink in self-adaptive Website by Web data minming. The first part of the paper is the introduction of the current status of developments in the field of Web Mining. Sencondly, it discusses the technologies of data mining and Web Mining, and explains how to implement them. Thirdly, it annalyzes and studis Web log mining. According to the above parts, the paper deeply explains the self-adaptive Website and realizes self-increased dynamic link, the paper. In the process of preprocessing, a user identification method based on cookie technology and extending Web Log attributes are adopted. The URL-UserID relevant matrix clustering algorithm is used for user access pattern mining. According to result of web usage mining, the program add automaticly dynamic hyperlink by inserting ASP code in web page to present different view to unlike user. In the last part of the paper is the conclusion, it gives a description of Web Mining and self-adaptive Website, and expects its application in the future.更多还原