挖掘关联规则中Apriori算法的改进与优化  

Improvement and Optimization of Apriori Algorithm in Mining Association Rules

作　　者：刘巍[1] 蒋华[1] 

机构地区：[1]桂林电子工业学院计算机系,广西桂林541004

出　　处：《计算机与现代化》2006年第11期113-115,共3页Computer and Modernization

摘　　要：在所有的关联规则的挖掘算法中Apriori算法是最为经典的一个,但Apriori算法有两个缺陷,即要扫描多次数据库以及生成大量的候选集。本文提出一种利用对项进行编码的方法对该算法进行改进,通过对项编码来减少扫描数据库次数并通过删除项来减少生成候选集的数量,从而提高算法的效率。相同条件下的实验结果表明,该优化后的算法能有效地提高关联规则挖掘的效率。Apfiofi algorithm is the most classical algorithm of all the association rules mining methods, but Apfiofi algorithm has two faults, firstly, this algorithm has to scan database many times , secondly, this algorithm has to produce many candidate set item. This paper improves this algorithm through making cede for every item. Making cede can reduce the times of scanning database,deleting items can reduce the number of candidate set items,so as to improve the efficiency of Apriori algorithm, The experiment results in the same environment show that the new algorithm can greatly improve the mining efficiency of association rules.

关 键 词：数据挖掘 关联规则 APRIORI算法 编码 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]