Read the full-text of this article for free

Title: The impact of a personalised music recommendation system driven by reinforcement learning on college students' psychological adjustment

Authors: Xiaomei Xu; Tiantian Xu

Addresses: Chinese Curricula Center, Wenzhou-Kean University, Wenzhou, 325000, China ' Chinese Curricula Center, Wenzhou-Kean University, Wenzhou, 325000, China

Abstract: Music serves as a convenient and effective tool for emotional regulation and holds significant value in psychological adaptation. Addressing the issue that existing research has overlooked real-time changes in college students' interests, leading to insufficient analysis of the impact on psychological adaptation, this study first embeds music input into a long short-term memory network model, modelling the sequence processing issue as the Markov decision process, and uses a multilayer perceptron model as the decision agent. High and low-score decision actions are input into a pseudo-twin network to generate delayed rewards. The agent gradually learns strategies that maximise rewards, enabling reliable music recommendations. Finally, the study analyses the systems impact on college students' psychological adaptation. Experimental results show that the proposed model's hit rate improves by at least 4.73%, significantly enhancing college students' psychological well-being.

Keywords: psychological adjustment; reinforcement learning; music recommendation; Markov decision process; long short-term memory network model.

DOI: 10.1504/IJICT.2025.150404

International Journal of Information and Communication Technology, 2025 Vol.26 No.44, pp.91 - 106

Received: 25 Sep 2025
Accepted: 25 Oct 2025
Published online: 12 Dec 2025 *