采用构造后缀树方法的在线挖掘浏览模式

在构建个性化服务中常常通过需要挖掘用户的浏览模式来改进站点的结构，使得客户访问站点变得容易。问题的关键就在于如何从Web日志中发现用户浏览模式。该文就此提出一种算法，通过为Web日志数据构造后缀树进行序列挖掘，找出最大频繁序列，进而发现有意义的浏览模式。该算法的重要特点是可以在线进行。

On-line Mining Traversal Patterns by Constructing Suffix-tree

When constructing personalized service, people often need mining the user's traversal patterns to improve the organization of the Web site and make it easier for clients to access. The key is how to find the traversal patterns from Web logs. To solve this problem, this paper proposse an algorithm by constructing a suffix-tree on Web-log, mines the maximal frequent sequence and find the interesting traversal patterns. The trait of this algorithm is that it can be operated online.