一种面向数据流的频繁项集挖掘算法  

An Algorithm for Mining Frequent Itemsets in Data Streams

作　　者：孟彩霞[1] 

机构地区：[1]西安邮电学院计算机科学系,陕西西安710061

出　　处：《昆明理工大学学报（理工版）》2009年第5期26-30,35,共6页Journal of Kunming University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金(项目编号:60573096);陕西省自然科学基金项目(项目编号:2004f283);西安市科技创新支撑-应用发展研究计划项目(项目编号:YF07024)

摘　　要：与传统静态数据库中的数据不同,数据流是一个按时间到达的有序的项集,这使得经典的频繁项集挖掘算法难以适用到数据流中.根据数据流的特点,提出了数据流频繁项集挖掘算法FP-SegCount.该算法将数据流分段并利用改进的FP-growth算法挖掘分段中的频繁项集.然后,利用Count Min Sketch进行项集计数.算法解决了压缩统计和计算快速高效的问题.通过和FP-DS算法的实验对比,FP-SegCount算法具有较好的时间效率.Different from data in traditional static database, a data stream is an ordered sequence of items that arrives in timely order. Classical frequent item - sets mining method is difficult to apply to data stream. Based on the characteristics of data streams, FP - SegCount algorithm is proposed in this paper to mine frequent item - sets from data streams. The algorithm partitions the data stream and uses modified FP - growth algorithm to mine frequent item- sets in every segment. It then counts item -sets in Count Min Sketch. This algorithm solves compressed statistics and ensures effective computation. Through experimentation and comparison with FP - DS algorithm, FP SegCount algorithm is shown to have a good time efficiency.

关 键 词：数据流 数据挖掘 数据流挖掘 频繁项集 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]