无支持度约束的靶向式关联规则挖掘

摘要

Some concepts such as all attribute itemset, absolute association rule, key antecedent of association rule are proposed to solve information annihilating problem caused by the combination explosive of itemset in associated rules mining without support. This paper proves an important theory, the association rule with the antecedent of key antecedent's super set must be absolute association rule, and it is upward closure. Based on this principle, a targeting association rule mining algorithm is designed to eliminate redundancy association rule significantly. Through an instance, the feasibility and effectiveness of the algorithm are verified.%为解决不考虑支持度时关联规则挖掘中数据项集组合爆炸引起的信息湮灭问题,给出全属性项目集、完全关联规则、关联规则的关键前提等概念.证明以关键前提的超集作为前提的关联规则也一定是完全关联规则,即向上闭合特性.根据该原理设计一个能够消除大量冗余关联规则的靶向式关联规则挖掘算法.通过挖掘实例验证了该算法的可行性和有效性.