基于关联规则的数据挖掘算法研究  

Research on Data Mining Algorithm s Based on Association Rules

作　　者：石正喜[1] 葛科奇[1] 曹财耀[1] 

机构地区：[1]宁波城市职业技术学院信息学院,浙江宁波315100

出　　处：《计算机与网络》2013年第6期62-64,共3页Computer & Network

摘　　要：在介绍数据挖掘、关联规则挖掘概念的基础上,详细分析了Apriori算法的特点及其不足之处。随后,分析了Apriori算法的改进思路,提出了改进的Apriori关联规则算法:基于可拓理论的Apriori算法。借助于实验数据,对经典的Apriori算法及改进的Apriori算法进行了对比分析研究,结果表明,通过Apriori算法挖掘出的规则含有大量的冗余规则,而改进的Apriori算法能够避免冗余规则的产生,挖掘出的结果集简洁且完备。Based on an introductionto the concept ofdata mining and association rules, the features and defects of Apriori algorithm are analyzed, the improvingideas of Apriori algorithm are analyzed, and an improved Apriori algorithm of association rules is proposed, that is, the Aprioti algorithm based on extension theory. The performances of classical Apriori algorithm and the improved Apriori algorithm arecompared by test data. The test results show that the hales minedby Apriori algorithm mining have many redundant rules, while the Apriori algorithm based on extension theory can avoid redundant rules and the mined rules are more simple and complete.

关 键 词：数据挖掘 关联规则 APRIORI改进算法 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]