利用组合支持度进行关联规则的挖掘  

Mining Association Rules with the Combination of Single Support and Multiple Supports

作　　者：阮璐[1] 肖冬荣[1] 周杰[1] 高风[1] 

机构地区：[1]南京信息工程大学信息与控制学院,江苏省南京市210044

出　　处：《微计算机信息》2008年第9期233-234,170,共3页Control & Automation

基　　金：江苏省高校自然科学研究指导性计划项目(05KJD510122)

摘　　要：经典Apriori算法由单支持度获取频集,这无法满足现实中的实际情况。国外学者提出的多最小支持度的扩展模型的计算次数又过于庞大,运行效率较低。针对这种情况,提出了多支持度与单支持度相结合的挖掘关联规则的方法,即保证了对不同数据项的正确分析,又提高了计算效率,有较强的实际意义。The classical Apriori algorithm gets the large item sets with the single support, and it can not satisfy the real situation. A foreign expert has developed an advanced model with the multiple supports based on the Apriori algorithm, but it didn t work very effectively. The combination of the single support and the multiple supports can solve these problems. Both the quality of the rules and the efficiency of the mining procession can be insured.

关 键 词：数据挖掘 关联规则 支持度 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]