基于Hash结构的关联规则交互挖掘算法

关联规则挖掘是数据挖掘的主要技术之一,现有的关联规则挖掘算法均基于支持度-置信度框架,当用户调整阈值时存在多次遍历数据库和重复计算问题。该文针对支持度阈值变化时的关联规则维护问题,提出了关联规则交互挖掘算法HIUA,该算法改进了原始IUA算法的剪枝过程,并通过Hash结构提高算法运行效率。在UCI数据集及企业实际财务数据集中的实验结果表明:在支持度阈值发生变化的过程中HIUA算法进一步利用已有挖掘结果,有效提高了关联规则挖掘的效率。

Interactive association rule mining algorithm based on a Hash structure

Association rule mining is one of the main data mining techniques,but most existing association rule mining algorithms are based on the support-confidence framework with most regressing multiple database scans and redundant computing when using a user adjusting support threshold.This paper presents a Hash-based IUA algorithm(HIUA) for association rule maintenance with support threshold changes.This algorithm improves the prone step in the original IUA and uses the Hash structure to improve the efficiency.Tests show that the HIUA effectively improves the association rule mining efficiency.