A study on the application of neural networks to determine students' mental health status in school education

Abstract

College students' studying and living environments have become more difficult as a result of modern society's growth, resulting in substantial psychological concerns. Most academics, on the other hand, are more interested in quantitative evaluations based on objective surveys and statistical analysis, or contrastive studies based on pre-designed intervention procedures. Few of them evaluate the combination of influencing elements when determining the severity of psychological distress. To address the issue, this research applies artificial neural networks (ANN) to the early detection and intervention of psychological crises among college students using physical education (PE). The psychological crises of college students was analysed with the structural equation, taking into account the interaction impact between affecting elements, using a suitable index system. Following that, a bidirectional long short-term memory (BiLSTM) network was built for the purpose of analysing college students' psychological crises, and the degree of psychological crisis was assessed using the emotional labelling attention mechanism. Finally, the impact of physical activity on the reduction of psychological stress in college students was thoroughly examined. Experiments confirmed the validity of the proposed approach.