一种基于滑动窗口的数据流频繁闭项集挖掘算法

频繁项集挖掘是数据流挖掘中的一个热点问题.提出了一种新的数据流频繁闭项集挖掘算法MFCI-SW.首先设计了两个新的数据结构:频繁闭项集表FCIL和频繁闭合模式树MFCI-SW-Tree,在此基础上以滑动窗口中的基本窗口为更新单位,在每个基本窗口中提取出频繁闭项集的数据项,将其支持度F和窗口序列号K存到FCIL中;然后随着新基本窗口的到来,通过删除频繁闭项集表中K值最小的数据项和插入新数据项完成对FCIL的更新和MFCI-SW-Tree树的裁剪;最后在MFCI-SW-Tree中可以迅速挖掘出满足用户需要的频繁闭项集.实验结果证明了该算法在执行效率上明显优于DS-CFI算法.

An Algorithm for Mining Frequent Closed Itemsets in Data Streams over Sliding Window

Frequent itemsets mining is a hot topic in data stream mining.A new algorithm,called MFCI-SW,is proposed for mining frequent closed itemsets in data streams in this paper.Firstly,frequent closed itemsets list(FCIL)and frequent closed itemsets tree by the sliding window(MFCISW-Tree)are introduced in MFCI-SW,and the sliding window iS divided into several basic windows.In each basic window,the data items in frequent closed itemsets are collected up,and the supporting degree F and the window sequence number K of them are stored in FCIL.Then,when a new basic window arrives,the pruning of MFCI-SW-Tree is completed by deleting the data item whose K is the least in FCIL and merging the new data items into FCIL.Finally,the frequent closed itemsets which satisfy the query requirements of users are output.Experimental results show that the execution efficiency of MFCI-SW is better than DS-CFI.