A mechanism for analyzing and managing undergraduates' mental health based on large‐scale behavior data: AI‐based approach

Abstract

AbstractManaging undergraduates' mental health is crucial for their well‐being and academic success. An artificial intelligence (AI)‐based approach using large‐scale undergraduates' behavior data can provide a mechanism for analyzing and managing undergraduates' mental health effectively. With the development of information communication technology in higher education, the daily behavior data of undergraduates is recorded in the college's database. Based on spending data in the canteen and time spent online on weekdays and weekends, Jenks natural breaks algorithm is first applied to classify features. Then, according to the feature classification results, the Apriori algorithm is used to analyze feature association to mine the behavioral characteristics that correlate with undergraduates' mental health. Finally, the inertia weight is improved based on the particle swarm optimization algorithm. Identifying, mutating, and selecting the inferior particles is added to avoid the algorithm falling into the local optimal solution. Simultaneously, a particle difference neural network model is constructed to predict undergraduates' mental health. Based on the experimental results from the undergraduates' behavior characteristics dataset, the proposed model has demonstrated superiority over traditional machine learning and deep learning models.