基于矩阵的模糊关联规则挖掘算法及其应用研究

Study and Application of Fuzzy Association Rule Mining Based on Matrix

针对布尔型关联规则不能表达挖掘对象中模糊信息的关联性,给出了一系列有关模糊关联规则的定义,并提出了一种基于矩阵结构的模糊关联规则数据挖掘算法(FARMBM)。该算法通过构造矩阵结构来压缩存储模糊模式候选集和频繁集,有效节约了存储模糊模式候选集和模糊模式频繁集内存花销,只需扫描数据库两遍,且可以有效减少系统的I/O开销。这里把FARMBM运用到入侵检测的仿真实验中,实验结果表明,该算法是有效的。

In allusion to the Boolean association rules can＇t express the association of fuzzy data, a series of definitions of fuzzy association rules and mining algorithm based on matrix for fuzzy association rules are proposed. The algorithm can store fuzzy pattern candidate sets and frequent sets compressible by constructing matrix structure, which effectively saves the memo- ry cost for storing fuzzy pattern candidate sets and frequent sets,it only scans database twice, besides it can effectively reduce the I/O spending. FARMBM is applied to the simulation results of intrusion detection,and efficiency of the algorithm is verified by the experiment.