Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (30): 20-22.DOI: 10.3778/j.issn.1002-8331.2010.30.006

• 博士论坛 • Previous Articles     Next Articles

Relationships between Web users as attributes for Web access patterns clustering

WU Rui   

  1. School of Mathematics and Computer,Shanxi Normal University,Linfen,Shanxi 041004,China
  • Received:2010-07-06 Revised:2010-09-14 Online:2010-10-21 Published:2010-10-21
  • Contact: WU Rui

以Web用户关联关系为属性的浏览模式聚类

吴 瑞   

  1. 山西师范大学 数学与计算机学院,山西 临汾 041004
  • 通讯作者: 吴 瑞

Abstract: Clustering Web access patterns in Web usage mining is an effective way to help Web site designers understanding Web users’ characteristic and needs.An effective method is proposed to cluster Web access patterns.Whether a Web page is visited or not and time duration on it disclose Web users’ interest.They are denoted by the corresponding item in vector of Web access pattern with the same length.Furthermore,relationships among Web access patterns are characterized and added into vector of Web access pattern.Then extended vector of Web access patterns are formed and clustered effectively by rough k-means.The clustering results are more reasonably by analysis of an example and experiment.They can be adopted to design personalized website.

Key words: Web access patterns, Web mining, clustering, rough set

摘要: 在Web使用挖掘中,用户浏览模式的聚类结果有助于网站设计者理解Web用户的浏览特点和需要。设计了一种有效的Web浏览模式的聚类方法,网页是否被浏览及网页上的浏览时间反映了用户的浏览兴趣,它们被刻画成等长的用户浏览模式向量中的相应分量,此外,浏览模式之间的关系被刻画并被作为属性加入到该向量中,形成扩展的用户浏览模式向量,对这些向量使用粗糙k-均值法可对用户浏览模式进行有效的聚类。实例和实验分析说明,使用该方法的聚类结果更合理。聚类结果可用于个性化网站的设计。

关键词: 用户浏览模式, Web挖掘, 聚类, 粗糙集

CLC Number: