多最小效用阈值的频繁高效用项集快速挖掘算法  

Fast mining algorithm for frequent and high utility itemsets with multiple minimum utility thresholds

作　　者：王斌 吕瑞瑞 房新秀 马俊杰 Wang Bin;Lyu Ruirui;Fang Xinxiu;Ma Junjie(School of Information&Control Engineering,Qingdao University of Technology,Qingdao Shandong 266033,China)

机构地区：[1]青岛理工大学信息与控制工程学院

出　　处：《计算机应用研究》2019年第12期3623-3627,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61502262)

摘　　要：针对多最小效用阈值高效用项集挖掘算法(MHUI)中出现的重复计算、挖掘的结果项集不是频繁的问题,提出两个新的快速挖掘算法FMHUI和SFMHUI。FMHUI算法在计算项集的最小效用阈值时利用前一次计算结果,避免了项之间的重复比较;另外定义了项的扩展项的最小效用阈值表EMMU-table快速计算出扩展项的最小效用阈值,提高了运行效率。SFMHUI算法在FMHUI的基础上增加了支持度约束,使挖掘的项集既是高效用的也是频繁的。通过仿真实验验证了所提出算法的高效性和可行性。In mining algorithm for high utility itemsets with multiple minimum utility threshold( MHUI),calculation is often repeated and mining result itemsets are not frequent. This paper developed two new fast mining algorithm SFMHUI and FMHUI. The FMHUI algorithm used the previous calculation result in the calculation of the minimum utility threshold of the itemsets to avoid duplicate comparisons between items. In addition,it defined the minimum utility threshold table EMMU-table of extensions of items to quickly calculate the minimum utility threshold of extensions,improved the efficiency. The SFMHUI algorithm added the support constraints on the basis of FMHUI,making the mining itemsets both high-utility and frequent. The result from simulation experiments shows that the proposed algorithms are efficient and feasible.

关 键 词：频繁项集 高效用项集 支持度 多最小效用阈值 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]