Frequent itemsets mining is a core data mining operation and has been extensively studied over the last decade. In the article, it overviews the techniques to mine frequent patterns over data streams. Then it gives some important definition, analyses some typical algorithms over data streams. Based on this, it summarises the meanings and theories from people before, and then it extends a new algorithm-FLY . After that, it analyses the efficient and accuracy of the new algorithm.%频繁模式挖掘是数据挖掘的精髓,并且在上个世纪被广泛研究.概述了数据流频繁模式挖掘的技术发展水平和发展背景.然后给出一些概念定义,就目前经典的数据流频繁模式算法作了分析,并在此基础上总结了前人的方法理论,采用混合的方法,扩展了新的算法——FLY.最后,总结分析新的算法,并与前人算法在内存消耗和准确性方面进行了比较.
展开▼