一种快速挖掘生成器算法  

A Fast Mining Generator Algorithm

作　　者：许普乐[1] 纪允[2] 

机构地区：[1]芜湖职业技术学院教务处,安徽芜湖241006 [2]中国移动通信集团安徽有限公司铜陵分公司,安徽铜陵244000

出　　处：《绍兴文理学院学报》2016年第7期63-68,共6页Journal of Shaoxing University

基　　金：安徽省高等教育振兴计划重大教学改革研究项目"职业院校信息化教学改革的研究与实践"(项目编号2014zdjy198);安徽省高校优秀青年人才支持计划重点项目"职业院校教育信息化发展路径研究-以安徽省为例"(项目编号:gxyqZD2016591)

摘　　要：生成器是频繁项集精简表示中的一个经典模型,但其传统挖掘算法存在重复生成候选项集,反复扫描数据库得到支持度,需要遍历所有直接子集等缺点,导致生成效率低下.基于此,一种快速挖掘生成器算法FMG,该算法采用Rymon枚举树作为搜索空间,提出的判断生成器定理对候选项集进行快速判断,以及特定的剪枝策略.通过这些方法快速的挖掘生成器.实验结果证明,该算法不仅比传统的算法要快,而且比最新提出的快速挖掘算法还要快.Generators are a classical model concisely represented in frequent itemsets, whose traditional miningalgorithm has such deficiencies as repeated generations of candidate itemsets and whose support obtained by re-peatedly scanning database needs traversing all direct subsets. All those drawbacks result in lower generation effi-ciency. Motivated by this, a fast mining generator algorithm FMG is proposed. This algorithm uses the Rymonsetenumeration tree as its searching space; a generator determiner lemma proposed in this paper is used to quick-ly determine a candidate itemset and to select particular paths for pruning. All the methods are helpful for mininggenerators. Experimental results show that this algorithm not only has better performance than traditional ones,but also is faster than a fast mining generator algorithm proposed recently.

关 键 词：数据挖掘 频繁项集 精简表示 Rymon枚举树 生成器 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]