Authors: Ming Wu
Addresses: College of Information and Electronic Engineering, Hunan City University, Yiyang, Hunan, 413000, China
Abstract: In order to overcome the problems of low cleaning efficiency and serious memory consumption in traditional large data cleaning methods, this paper proposes a new cleaning method of repeated big data based on association rule mining algorithm. This method uses association rule mining algorithm to obtain the frequent itemsets of repeated big data after repeated cycle calculation. At the same time, the output mode of the algorithm is optimised in parallel, and the Hadoop interface is modified to change the reading mode of MapReduce. The first frequent itemset is used to clean the repeated big data. The experimental results show that the proposed method can effectively reduce the execution time and memory consumption, and the shortest cleaning time is only 1.28 min, indicating the feasibility of the proposed method.
Keywords: low cleaning efficiency; serious memory consumption; association rule mining algorithm; duplicate big data; cleaning; frequent items.
International Journal of Autonomous and Adaptive Communications Systems, 2023 Vol.16 No.2, pp.220 - 231
Received: 17 Apr 2020
Accepted: 03 Sep 2020
Published online: 24 May 2023 *