摘要
针对已有两阶段高效用挖掘算法在遍历解空间时耗时过长的问题,提出一种随机高效用模式挖掘算法。即在阶段一置若干随机数,每个随机数对应一个节点,随后计算该节点的事务权重效用值并利用事务权重向下闭包的特性,若该节点的事务权重效用值低于设定阈值,则该项集的任意超集被剪枝。实验表明,该算法平均运行效率相较原始算法有明显提升。
Aiming at the problem that the existing two-stage high utility mining algorithm takes too long to traverse the solution space, a random high utility pattern mining algorithm is proposed. That is, set several random numbers at one stage, and each random number corresponds to a node. Then calculate the transaction weight utility value of the node, and use the transaction weight downward closure feature. If the transaction weight utility value of the node is lower than the set threshold, any superset of the itemset will be pruned. Experiments show that the average running efficiency of the algorithm is significantly improved compared with the original algorithm.
作者
钟新成
李慧芳
ZHONGXin-cheng;LIHui-fang(DepartmentofComputerScience,ChangzhiUniversity,ChangzhiShanxi,046011)
出处
《山西大同大学学报（自然科学版）》
2022年第2期21-23,共3页
Journal of Shanxi Datong University(Natural Science Edition)
基金
长治学院科研项目[020-XN0054]。
关键词
高效用
事务权重效用
剪枝
high utility
transaction weighted utility
prune
