Apriori algorithm based prediction of students' mental health risks in the context of artificial intelligence.

Introduction: The increasing prevalence of mental health challenges among college students necessitates innovative approaches to early identification and intervention. This study investigates the application of artificial intelligence (AI) techniques for predicting student mental health risks. Methods: A hybrid predictive model, Prophet-LSTM, was developed. This model combines the Prophet time series model with Long Short-Term Memory (LSTM) networks to leverage their strengths in forecasting. Prior to model development, association rules between potential mental health risk factors were identified using the Apriori algorithm. These highly associated factors served as inputs for the Prophet-LSTM model. The model's weight coefficients were optimized using the Quantum Particle Swarm Optimization (QPSO) algorithm. The model's performance was evaluated using data from a mental health survey conducted among college students at a Chinese university. Results: The proposed Prophet-LSTM model demonstrated superior performance in predicting student mental health risks compared to other machine learning algorithms. Evaluation metrics, including the detection rate of psychological issues and the detection rate of no psychological issues, confirmed the model's high accuracy. Discussion: This study demonstrates the potential of AI-powered predictive models for early identification of students at risk of mental health challenges. The findings have significant implications for improving mental health services within higher education institutions. Future research should focus on further refining the model, incorporating real-time data streams, and developing personalized intervention strategies based on the model's predictions.