| 摘要: |
| 提出了结合站点拓扑结构和Web页面内容的页面聚类改进算法,改进算法引入Web页面的内容链接比和页组的组内链接度,并修改了频繁访问页组支持度的计算公式,以此来提高挖掘结果的兴趣性.通过实验数据的比较,改进算法较一般算法的收敛性好,发现的频繁访问页组的兴趣性高. |
| 关键词: Web日志挖掘 日志分析 页面聚类 频繁访问页组 |
| DOI： |
| 分类号: |
| 基金项目:上海市科技发展基金资助项目(985115035) |
|
| Mining the Page Clustering Based on the Content of Web Pages and the Site Topology |
|
YANG Yi-ling,GUAN Xu-dong,YOU Jin-yuan
|
| Abstract: |
| In this paper, an enhanced algorithm is proposed for page clustering, which considers both the content of web pages and the site topology. By introducing the content-link ratio and the group inter-link degree and modifying the computation of the support of frequently visited page group, the algorithm can increase theinterestingness of the mining result. The experimental results show that the algorithm converges more rapidly and could find out more interesting page groups than the normal algorithm. |
| Key words: Web log mining log analysis page clustering frequently visited page group |
