Advanced Data Analysis and Prediction Model for Student Mental Health Risk Assessment

Xiuyu Shi^1*Jin Pan²Daofu Yuan³Minye Li²Yafeng Pan⁴

• ¹Other
• ²Zhejiang Institute of Communications, Hangzhou, China
• ³Zhejiang Normal University, Jinhua, China
• ⁴Zhejiang University, Hangzhou, China

With the increasing prevalence of mental health issues among students, early detection plays a vital role in ensuring timely intervention and support. However, existing methods often face challenges in effectively capturing the complex relationships among diverse data sources, such as behavioral, emotional, and physiological data, and in considering the temporal dynamics of mental health changes. To address these limitations, we propose PsyGraph-SSL, a novel model that combines graph convolutional networks (GCN), temporal modeling, and self-supervised learning (SSL) to predict and analyze student mental health risks. By integrating multi-modal data and learning temporal dependencies, PsyGraph-SSL offers a more accurate and comprehensive solution for mental health prediction. Experimental results on the WESAD and Student Well-Being Dataset show that PsyGraph-SSL outperforms traditional models in terms of accuracy, F1 score, AUC, and other key metrics. The model excels at capturing emotional and behavioral fluctuations, demonstrating its strong potential for early detection and intervention. This study presents an innovative framework for student mental health monitoring, emphasizing the importance of multimodal data fusion and temporal analysis, with the potential to contribute to the development of real-time, adaptive mental health intervention systems.