Abstract: In order to solve the shortcomings of association rule mining algorithms in Web personalization，especially that it can not reflect the change of user’s interest in time，this paper proposes an algorithm for dynamic adaptive Web site based on association rule mining（DPARM）.DPARM combines both recommendation and user profile updating，it updates the usage profiles of customers as soon as possible in order to make the personalization system can reflect the diversions of customers’ interest.Experiments have been done with the usage data of http：//www.cs.depaul.edu，and the experiments show that DPARM is effective and practical.

Key words: association rule, Web personalization, online recommendation

摘要： 针对当前个性化中关联规则挖掘的一些问题，尤其是无法及时更新使用数据这一缺点，提出了一种有效的基于关联规则挖掘的个性化方法DPARM，它将用户兴趣模型的更新和在线推荐紧密结合，及时使用新的用户会话更新用户兴趣模型，从而使个性化系统能够更好反映用户访问模式的变换。使用http：//www.cs.depaul.edu上的数据进行了实验，结果表明，该方法是有效可行的。

关键词: 关联规则, Web个性化, 在线推荐