数据采集与处理2012，Vol.27Issue(4)：508-513,6.
在线挖掘数据流闭合频繁项集CMNL-SW算法
CMNL-SW Algorithm on Online Mining Closed Frequent Itemsets Over Data Stream
汤春明 1王培义 1曲英涛2
作者信息
- 1. 哈尔滨工程大学信息与通信工程学院,哈尔滨,150001
- 2. 哈尔滨工程大学网络信息中心,哈尔滨,150001
- 折叠
摘要
Abstract
A new online mining algorithm called the closed map and num list-sliding window (CMNL-SW) is proposed. It uses two data structures, i. e. closed map stores, the closed item-sets, those are mined and the num list stores the number of all different items. Via the simple union operation on item number contained within a new arriving or an old deleting transaction and the intersection operation on certain previous closed itemsets once, it incrementally updates the current sliding window and makes the closed frequent itemsets be output in real time based on the specified thresholds of any user. Theoretical analysis and experimental results of the real datasets, such as mushroom, retail-chain and artificially synthesized datasets T40I10D100K show that the proposed method is superior to the classic algorithms Moment and CFI-Stream in terms of time and space efficiencies, and it has good stability as the number of transactions processed increases and adapts rapidly to the change in data streams. 关键词
挖掘算法/闭合频繁项集/滑动窗口/数据流Key words
mining algorithm/ closed frequent itemsets/ sliding window/ data stream分类
信息技术与安全科学引用本文复制引用
汤春明,王培义,曲英涛..在线挖掘数据流闭合频繁项集CMNL-SW算法[J].数据采集与处理,2012,27(4):508-513,6.
