Title: Enhancing user intimacy: a parallel mining algorithm utilising UISFW

Authors: Ruiping Kong; Ruimin Pan

Addresses: Department of Intelligent Technology, Tianjin Polytechnic College, Tianjin, 300400, China ' School of Information Technology and Engineering, Tianjin University of Technology and Education, Tianjin, 300500, China

Abstract: User intimacy is vital for gauging the emotional bond and interaction between users and brands, products, or services. Existing algorithms usually struggle with accuracy and potential relationship identification. To address these issues, the user intimacy is calculated based on intimacy feature assignment. Genetic algorithm is used to complete feature assignment. It can efficiently solve complex optimisation problems and quickly find approximate optimal solutions. Then, MapReduce is used to construct a parallel mining model for user intimacy, which can simplify the complexity of parallel computing and data processing. The combination of the two can optimise data processing efficiency. The results showed that the proposed method had average accuracy (P@n) of 90.28% and 90.89%, and normalised discounted cumulative gain (NDCG) values of 82.88% and 82.73% across datasets. The model effectively computes user intimacy and uncovers valuable insights for user relationship identification, benefiting product design and information recommendations.

Keywords: user intimacy; feature assignment; genetic algorithm; parallel mining; MapReduce.

DOI: 10.1504/IJWET.2025.151160

International Journal of Web Engineering and Technology, 2025 Vol.20 No.4, pp.441 - 464

Received: 29 Apr 2024
Accepted: 06 Mar 2025
Published online: 15 Jan 2026 *

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