Title: A social network security user recommendation algorithm based on community user emotions

Authors: Huajin Liu; Chunhua Ju; Houyuan Zhang

Addresses: School of Business Administration, Zhejiang Gongshang University, Hangzhou 310018, China; School of Network Communication, Zhejiang YueXiu University, Shaoxing 312000, China ' School of Business Administration, Zhejiang Gongshang University, Hangzhou 310018, China ' College of Humanities and Arts, Ningbo University of Technology, Ningbo 315211, China

Abstract: Social networks play a vital role in people's lives and work, but have problems with sparse data and cold start. This study establishes a social network model and innovatively improves the classic user interest point recommendation algorithm based on community information and user emotion. A sequential learning ranking algorithm is designed to simulate user preferences from a sequence of recommended objects and convert user ratings into ranking scores, combined with a network security dictionary, Node2vec method, and hot coding to capture network security vocabulary. This study also uses the heuristic firefly optimisation algorithm to solve the problem and confirms that community CU-SNR has good experimental results. The improved LDA algorithm is used to adjust the social media emotion data, and three real social network data sets verify the algorithm's performance. Numerical experiment results show that the algorithm simulation has a certain effect when facing social networks.

Keywords: community information; user characteristics; social networks; firefly algorithm; recommendation algorithm; linear discriminant analysis; LDA.

DOI: 10.1504/IJSN.2024.137332

International Journal of Security and Networks, 2024 Vol.19 No.1, pp.10 - 19

Received: 14 Sep 2023
Accepted: 24 Sep 2023
Published online: 12 Mar 2024 *

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