基于距离的关联规则相关性分析优化方法

Distance-based optimization approach for correlation analysis of associ- ation rule mining

关联规则挖掘常常会产生大量的规则,这使得用户分析和利用这些规则变得十分困难。为了帮助用户做探索式分析,提出了一种基于距离的相关性关联规则优化方法,该方法从数学分析关联规则相关性概念公式的值的特点出发,通过根据关联规则结构上的相关性差别来挖掘出包括正负两种关联规则在内的更多潜在的相关规则,实验结果表明该方法有效且可靠。

A common problem in association rule mining is that a large number of rules are often generated from database.It makes users difficult to analyze and use these rules.To facilitated exploratory analysis,a distance-based optimization approach is used for correlation analysis of association rule mining.The approach mathematically analyzes the value of correlation formulation between correlation rules.Positive and negative rules,as well as other rules can be distinguished and extracted by analyzing the difference of correlation rules.The experiments demonstrate that the proposed approach is effective and credible.