关联规则衡量标准的研究-控制与决策.PDF

第18 卷第3 期 　控　制　与　决　策 2003 年5 月 Vol. 18 No.3 　Control and Decision　 　May 2003 : 1001-0920(2003) 03-0277-04 1, 2 1 罗　可 , 吴　杰 ( 1. , 410082; 2. , 410077) : 关联规则采掘是数据采掘中重要的研究课题。针对当前关联规则采掘中可能产 许多无效关 联规则的问题, 分析其原因, 提出在衡量标准中增加有效度, 并给出了有效度的定义。根据有效度的大 小,将关联规则分为正关联规则、无效关联规则、负关联规则, 提出了新衡量标准采掘关联规则的算法, 并用Visual FoxPro 进行了试验。实验表明,新方法能明显减少无效关联规则的数目。 : 数据采掘; 关联规则; 有效度; 算法 : TP311: Research on j udgment criterion of association rul es 1, 2 1 L UO K e , WU Jie ( 1.School of Electric and Information Engineering, Hu nan University, Changsha 410082, China; 2.Division of Computer pplication Changsha University of Electric Power Changsha, , 410077, China) A bstract: Mining association rules are an important topic in the data mining research The reasons for. many invalid rules in mining association rules are analyzed. The validity is defined and added to the judgment criterion. ccording to the value of validity, association rules are classified into positive, invalid and negative association rules. n algorithm of new judgment criterion in mining association rules is presented and tested with Visual FoxPro The test results show that the method can obviously. reduce invalid association rules. Key words : Data mining; ssociation rules; Validity; lgorithm 1引　　 言 s,sD X Y, (Transaction) P (X Y) ;2) c,c D X , Y , P( YX) [ 1] D grawal : I = {i1,i2, ,im} m minsup ,D I , minconf ,, ii ,ij , ,ik