不确定数据频繁项集挖掘算法研究  

Research on Frequent Itemset Mining Algorithm for Uncertain Data

作　　者：赵学健[1] 熊肖肖 张欣慧[2] 孙知信[1] ZHAO Xue-jian;XIONG Xiao-xiao;ZHANG Xin-hui;SUN Zhi-xin(School of Modern Posts,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;School of Internet of Things,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]南京邮电大学现代邮政学院,江苏南京210003 [2]南京邮电大学物联网学院,江苏南京210003

出　　处：《计算机技术与发展》2019年第7期140-144,共5页Computer Technology and Development

基　　金：国家自然科学基金(61373135,61672299);国家自然青年科学基金(61702281,20140883);江苏省基础研究计划(自然科学基金)(BK20140883,BK20140894,BK20150869)

摘　　要：频繁项集挖掘的目标是以频繁出现的项目集的形式发掘嵌入在海量数据中的隐式的、先前未知的、潜在的有用知识,以辅助决策。随着数据采集方式和传输方式的多样化,不确定数据在各种实际应用中大量出现。因此,近年来针对不确定数据的频繁项集挖掘算法的研究引起了学者的广泛关注。文中首先介绍了不确定数据的定义,并分析了不确定数据频繁项集挖掘的概率模型。接下来,将主流频繁项集挖掘算法分为3类:基于候选项集生成和测试的频繁项集挖掘算法,基于模式增长的频繁项集挖掘算法和基于生物启发的频繁项集挖掘算法,详细介绍了当前针对不确定数据的主流频繁项集挖掘算法,并对这些算法的性能进行了简单分析。最后,对不确定数据的频繁项集挖掘算法进行了总结与展望。The frequent itemset mining aims to explore the implicit,previously unknown and potential useful knowledge embedded in big data in the form of frequent itemsets to assist the decision-making.With the diversification of data acquisition mode and transmission mode,uncertain data appear in a large number of practical applications.Therefore,in recent years,the research on frequent itemset mining algorithm for uncertain data has attracted wide attention from scholars.In this paper,we first introduce the definition of uncertain data and analyze the probability model of mining frequent itemset for uncertain data.Then,we divide the typical frequent itemset mining algorithms into 3 categories:candidate generate-and-test based frequent itemset mining algorithms,pattern growth based frequent itemset mining algorithms and bio-inspired frequent itemset mining algorithms.Typical frequent itemset mining algorithms are introduced.Moreover,the performance of these algorithms is also analyzed.Finally,the algorithm of mining frequent itemsets for uncertain data is summarized and prospected.

关 键 词：频繁项集 不确定数据 候选项集 模式增长 生物启发 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]