摘要： 针对频繁项集挖掘时间与空间效率低的问题，提出一种基于前缀树的高效频繁项集挖掘算法，通过对事务集进行预处理，创建索引表并分配索引编号，保证前缀树中事务顺序的一致性，根据索引编号等信息创建紧凑的前缀树，采用自底向上的挖掘与投影的方式挖掘出频繁项集。实验结果表明，该算法挖掘效率高、占用空间少。

关键词: 频繁项集, 数据挖掘, 前缀树

Abstract: Aiming at the problem of low time and space efficiencies for frequent item sets mining, an efficient frequent item sets mining algorithm based on prefix tree is proposed. To ensure the consistence of transactions sequence, the proposed algorithm pre-processes transaction sets to create index table and assign index identity. It creates compact prefix tree with the index information, and mines frequent item sets by bottom-to-up and projection methods. Experimental results show this algorithm has higher mining efficiency and expends less space.

Key words: frequent item set, data mining, prefix tree

中图分类号: 

• TP312