基于粗糙集的关联规则挖掘方法

Approach for mining association rules based on rough set

对粗糙集进行了相关研究,并提出一种以粗糙集理论为基础的关联规则挖掘方法,该方法首先利用粗糙集的特征属性约简算法进行属性约简,然后在构建约简决策表的基础上应用改进的Apriori算法进行关联规则挖掘。该方法的优势在于消除了不重要的属性,减少了属性数目和候选项集数量,同时只需一次扫描决策表就可产生决策规则。应用实例及实验结果分析表明该方法是一种有效而且快速的关联规则挖掘方法。

In this paper a novel approach for mining association rules based on rough set theory was presented. It firstly used the rough set＇ s feature attributes reduction algorithm to reduce attributes, and then applied the improved Apriori algorithm to mine association rules based on the reduction decision table. The advantages of this method were that it reduced the number of the attributes by reducing the unimportant attributes, also reduced the number of candidate item sets and could produce the association rules by scanning decision table only once. The results of application and experiments show that this method is fast and effective.