一种用户行为模式的挖掘算法

User Behavior Pattern Mining Algorithm

最大频繁事件序列挖掘是数据挖掘中重要的研究课题之一。该文提出了一种新的挖掘用户行为模式的算法。该算法采用位图索引表的数据格式,使用一种有效的基于前缀树的频繁事件序列扩展方法,结合有效的剪枝技术,明显地加速了最大频繁事件序列的生成。

Mining maximal frequent event sequence is an important topic in the data mining research. A new algorithm for mining user behavior pattern is proposed. It adopts a data format of bitmap index, an effective method based on extension of the frequent event sequence of the prefix tree, and effective prune techniques which prune the search space. Experimental comparison indicates that it accelerates the generation of maximal frequent event sequence obviously.