A new closed frequent itemset mining algorithm based on GPU and improved vertical structure. (2nd August 2016)

• Record Type: Journal Article
• Authors: Li, Yun; Xu, Jie; Yuan, Yun‐Hao; Chen, Ling
• Other Names: Fox Geoffrey guestEditor.; Dong Fang guestEditor.; Luo Junzhou guestEditor.
• Is Part Of: Concurrency and computation. Volume 29:Number 6(2017)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2016-08-02
• Subjects: data mining -- closed frequent itemset -- vertical data structure -- GPU; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.3904
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: Vertical data structure is very important for closed frequent itemset mining. All closed frequent itemsets can be found by simply using the operations of AND/OR. However, it consumes a large amount of storage space, especially in the case of large‐size dataset. This paper proposes an algorithm for mining closed frequent itemsets based on a new vertical data structure. The proposed data structure is helpful to save storage space by using a multi‐layer index. At the same time, numerous CPU and graphics processing unit can be employed in parallel to achieve high‐efficiency computing. Especially when dealing with large datasets, the proposed algorithm can obtain a high‐speed computing with the help of graphics processing unit. The improved vertical structure reduces the storage space of the data. The experimental results show that our proposed algorithm requires much less computation time than other related methods. Copyright © 2016 John Wiley & Sons, Ltd.