• 论文 •    

基于用户兴趣的Web日志挖掘算法

刘慧君,朱庆生,张程,周明强   

  1. 重庆大学 计算机学院,重庆400044
  • 出版日期:2009-11-15 发布日期:2009-11-25

Web log mining algorithm based on user interest

LIU Hui-jun, ZHU Qing-sheng, ZHANG Cheng, ZHOU Ming-qiang   

  1. College of Computer Science, Chongqing University, Chongqing 400044, China
  • Online:2009-11-15 Published:2009-11-25

摘要: 为有效地从Web日志中挖掘出用户访问模式,在分析用户访问行为的基础上,充分考虑用户在路径选择及在页面浏览中表现出的兴趣,提出了路径选择兴趣度以及页面浏览兴趣度的概念。设计了路径选择兴趣矩阵和页面浏览兴趣矩阵,并基于两个矩阵,设计了浏览兴趣路径挖掘算法。该算法先通过对两个矩阵进行路径兴趣度计算得到所有的浏览兴趣子路径,然后进行子路径合并生成浏览兴趣路径。实验表明了该算法的有效性。

关键词: Web日志, 浏览兴趣路径, 兴趣度, 挖掘算法, 电子商务

Abstract: To mine users'access pattern from Web logs effectively, based on analyzing users'access activities, the path-selection interest measure and the page-browsing interest measure were proposed by considering interests revealed both in path selection and page browse. After the definition of these two concepts, a browsing interest paths mining algorithm was devised according to the path-selection interest matrix and the page-browsing interest matrix. These two matrices were firstly set up from Web logs in the light of users'path-selection and page-browsing interest, and then browsing interest sub-paths were computed by the matrices. Finally, browsing interest paths were obtained by combination of all the sub-paths. Effctiveness of this proposed algorithm was proved by experiments.

Key words: Web log, browsing interest paths, interest measure, mining algorithm, e-business

中图分类号: