Abstract: The traditional algorithms simply regard frequent access paths as traversal interest paths，it’s not accurate.User’s interest page is mined through Web log by the navigation times，browsing time and information amount.The concept of user traversal interest and the algorithm for mining interesting navigation patterns is proposed.Experiments show the algorithm is feasible and effectual，and it is more accurate than other traditional algorithms of frequent access paths in measuring user traversal interests.

Key words: user traversal interest, user traversal pattern, data mining, sequential pattern mining

摘要： 针对当前挖掘用户访问模式算法仅将频繁访问路径作为用户浏览兴趣路径的问题，依据使用Web日志挖掘用户兴趣页面时，通过引入页面信息量参数，综合考虑页面访问次数、浏览时间和页面信息量大小来定义用户兴趣度，提出了基于兴趣度的用户访问模式挖掘算法。实验证明该算法是有效的，在用户浏览兴趣度量方面比当前的频繁访问路径挖掘算法更准确。

关键词: 用户浏览兴趣度, 用户访问模式, 数据挖掘, 序列模式挖掘