基于统计相关性的兴趣关联规则的挖掘

Correlation-Based Interestness Association Rules Mining

本文首先对关联规则的支持—置信框架存在的不足进行了分析,然后引入了规则的兴趣度概念,利用兴趣度来约束冗余关联规则的产生,以提高挖掘知识的有用性,并给出了算法描述。

In this paper, we analyze some problems existing in those available association rules mining algorithms, and then introduce a correlationbased interestingness measure. With this interestingness measure , more interesting rules can be mined. 