基于加权动态树的高权重容错频繁项集挖掘算法  

High weight fault-tolerant itemset mining algorithm based on weighted dynamic tree

作　　者：王斌 刘昊 李晓华 周伟 胡克勇 WANG Bin;LIU Hao;LI Xiaohua;ZHOU Wei;HU Keyong(School of Information and Control Engineering,Qingdao University of Technology,Qingdao 266525,China)

机构地区：[1]青岛理工大学信息与控制工程学院,青岛266525

出　　处：《青岛理工大学学报》2023年第3期130-137,共8页Journal of Qingdao University of Technology

基　　金：国家自然科学基金资助项目(61902205)。

摘　　要：针对传统频繁项集挖掘不能挖掘隐式频繁项集导致的输出频繁项集数量少、有潜在价值的项集未被发现、无法从利润方面体现数据价值等问题,提出了一种基于加权动态树的高权重容错频繁项集挖掘算法(HWFT-WDT)。该算法用于挖掘高权重容错频繁项集,保证用户能获得更加完整的项集及重要性信息;提出了加权动态树的数据结构,能够保存每个节点的权重,便于平均权重的计算;仅使用一个加权动态树的方法,避免了构造多个子树的高成本;提出了3个剪枝策略,有效地缩小了挖掘过程中的搜索空间。实验结果表明,该算法在运行时间、存储空间及延展性方面皆优于FT-PatternGrowth算法及FT-Apriori算法。Aiming to solve the problems that traditional frequent itemset mining cannot mine implicit frequent itemset,the number of output frequent itemset is small,potentially valuable itemset fail to be discovered and the data value cannot be reflected from the perspective of profit,this paper proposed a high weight fault-tolerant frequent itemset mining algorithm based on weighted dynamic tree(HWFT-WDT).This algorithm can be used to mine high weight fault-tolerant frequent itemset and ensure that users could obtain complete itemset and important information.This paper proposed a weighted dynamic tree data structure,which could store the weight of each node and facilitate the calculation of average weight.Only one weighted dynamic tree was used to avoid the high cost of constructing multiple subtrees.The algorithm proposed three pruning strategies,which effectively reduced the search space during the mining process.The experimental results show that the algorithm is superior to FT-PatternGrowth algorithm and FT-Apriori algorithm in terms of running time,storage space and scalability.

关 键 词：数据挖掘 加权动态树 容错模式 高权重 剪枝策略 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]