摘要
近年来,关联规则的挖掘已成为数据挖掘技术的重要研究方向之一。而对关联规则的研究也集中在如何高效的找出频繁项集。本文提出了一种基于Trie的在可信度构架下进行关联规则挖掘的方法,用于解决支持度为零的一类特殊问题,在不生成候选集的基础上,直接计算出所有的子集,节省了生成频繁项集的时空开销。
In recent years, mining association rules has became a most important aspect of research of Data Mining. And many researches of association rules focus on how to find out frequent itemsets. This paper proposed an association rules mining method based on Trie. It can be used to solve the problems with the zero support. This method direct calculation of all subset without generate candidate itemsets so that it saving time and space in generation of frequent itemsets.
出处
《心智与计算》
2010年第1期63-70,共8页
Mind and Computation
