Title: An optimal matching algorithm for e-commerce recommendation information based on matrix decomposition

Authors: Liming Wang; Wenxue Liu; Huichuan Liu

Addresses: Shandong Labor Vocational and Technical College, Ji Nan, 250022, China ' Shandong Labor Vocational and Technical College, Ji Nan, 250022, China ' Shandong Labor Vocational and Technical College, Ji Nan, 250022, China

Abstract: Aiming at the problems of data sparsity and cold start in traditional e-commerce recommendation information matching, an optimal matching algorithm for e-commerce recommendation information based on matrix decomposition is proposed. On the basis of e-commerce behaviour, the optimal user behaviour is obtained by decomposing the user's behaviour and the user's preference, and then the optimal user behaviour is obtained by decomposing the user's behaviour and the user's preference. The experimental results show that compared with the traditional optimal matching algorithm for recommendation information, the proposed optimal matching algorithm for e-commerce recommendation information based on matrix decomposition has a higher area under curve (AUC) value and a lower root-mean-square error (RMSE) value, and the performance of recommendation information matching is better.

Keywords: matrix decomposition; e-commerce; recommendation information; matching algorithm; hamming distance.

DOI: 10.1504/IJAACS.2023.134845

International Journal of Autonomous and Adaptive Communications Systems, 2023 Vol.16 No.6, pp.597 - 610

Received: 02 Sep 2020
Accepted: 10 Oct 2021
Published online: 14 Nov 2023 *

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