带约束的负关联规则挖掘算法

Algorithm for mining constrained negative association rules

针对仅有的挖掘算法不能较好地解决负关联规则的候选集数量爆炸问题,为满足用户的实际需求,提出带约束负关联规则概念,建立带约束负关联规则挖掘算法CNARM.同时,在挖掘过程中,利用最大频繁模式的性质来生成候选集,通过限制负关联规则中的前后件项目个数和利用负关联规则的性质来缩小候选集的规模.理论分析和实验结果表明本文提出的算法是有效可行的,具有较好的挖掘效率.

The scanty algorithms for mining can＇ t solve the problem of an exploding number of candidates well. So, an algorithm CNARM to satisfy the users＇ needs, which introduces a concept of constrained negative association rule, was proposed. At the same time, the character of maximal frequent patterns was used to generate candidates, by using the character of negative association rules, the upper bound of the former and back of negative association rule were restricted to reduce the size of candidates. Theoretical analysis and experimental results show that the algorithm proposed in this thesis is effective and feasible, and has better efficiency in the mining process.