数据流环境下的频繁项集挖掘综述

近些年来,在数据流环境下进行数据挖掘已得到该领域的热点关注。数据流与静态数据有很大不同,它具有无限、连续、高速和实时等动态特征,这就使得以往传统的频繁项集挖掘算法不在适用数据流环境。因此,本文将针对数据流环境下的频繁项集挖掘进行研究,将其分为三个方面,分别对其代表算法进行详细分析,做出对比并进一步总结,为学者们呈现出数据流挖掘在过去十余年中的发展,同时总结现有研究中存在的不足,提出未来可能的研究方向。

In recent years,data mining in the data flow environment has attracted much attention in this field.Data flow is very different from static data.It has such dynamic characteristics as infinite,continuous,high speed and real time,which makes the traditional frequent itemset mining algorithm not suitable for data flow environment.Therefore,this article will focus on data flow under the environment of frequent itemsets mining,which can be divided into three aspects,the detailed analysis of its representative algorithms respectively,make a comparison and summary,further present a data stream mining for scholars in the past more than ten years of development,summarizes the existing shortcomings in the course of study at the same time,the future research direction is put forward.