一种改进的Apriori算法
首发时间:2010-07-30
摘要:关联规则挖掘是指从一个大型的数据库中发现有趣的关联或相关关系,即从数据集中识别出频繁项集,然后利用这些频繁项集创建描述关联规则的过程。Apriori算法主要有两个操作,即连接步和剪枝步。本文提出的改进算法,对产生的频繁项集按字典顺序排列,在连接操作时减少连接的次数;对数据库的扫描操作转化为对矩阵的运算,同时对映射成的矩阵按项目的个数升序排列,以减少矩阵运算的运算量。实验表明,改进后的算法较之Apriori算法在时间性能上有明显的提高。
For information in English, please click here
Improvement on Apriori algorithm
Abstract:Association rules mining is used to mine interesting relations and relevance from huge data sets or database. In other words, it is to find out frequent itemsets and describe the association rules using the frequent itemsets. Apriori algorithm composes of two steps:connection step and deletion step. An improved method is presented in this paper. Firstly, in order to cut the count of connection, the frequent itemsets is order by alphabet. Secondly, this method not only uses matrix operation instead of the query operation in database, but also oder the matrix by the number of items contained in each transaction. An experiment shows that this improved method has high performance in obtaining bonus time of calculating.
Keywords: algorithm theory data mining Apriori algorithm association rules
基金:
论文图表:
引用
No.4379837538309127****
同行评议
共计0人参与
勘误表
一种改进的Apriori算法
评论
全部评论