Title: Research on fast mining algorithm for multi-feature fuzzy association data based on compressed matrix

Authors: Yibing Han; Zhanlei Shang

Addresses: Engineering Training Center, Zhengzhou University of Light Industry, Zhengzhou, 450001, China ' Engineering Training Center, Zhengzhou University of Light Industry, Zhengzhou, 450001, China

Abstract: In order to overcome the low mining accuracy and efficiency of traditional multi-feature fuzzy association data mining algorithms, a new fast multi-feature fuzzy association data mining algorithm based on compressed matrix is proposed in this paper. The compressed matrix structure is used to compress the fuzzy correlation data and generate the learning and training module. The average weighting method is used to extract fuzzy features, and the rule information of association data is integrated according to the mining mechanism to obtain the weighted confidence of association rules of fuzzy data. After data weighting, the optimal solution of fuzzy association rules is finally obtained, and the fast mining of fuzzy association data is completed. The experimental results show that the algorithm has accurate data mining effect, the execution speed of the algorithm is fast, and the maximum mining time is only 5.7 s.

Keywords: compression matrix; data support degree; membership degree; association rules; data mining.

DOI: 10.1504/IJICT.2024.137933

International Journal of Information and Communication Technology, 2024 Vol.24 No.3, pp.273 - 288

Received: 02 Mar 2021
Accepted: 31 Aug 2021
Published online: 11 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article