UNIVERSITY STUDENTS MENTAL STRESS DETECTION

Abstract—University students' mental stress has become a serious issue, affecting their general quality of life, well-being, and academic achievement. Several machine learning and deep learning techniques have been investigated to identify students' stress levels using textual, behavioural, and physiological data. However, advances in artificial intelligence (AI), machine learning (ML), and deep learning (DL) have led to the development of objective and data-driven stress detection methods. To accurately classify stress levels, physiological signals such as heart rate variability (HRV), galvanic skin response (GSR), and electroencephalography (EEG) were combined with behavioural and textual data from speech and social media posts. This paper provides a comprehensive review of existing stress detection techniques, with a particular emphasis on wearable sensor-based monitoring, speech and text analysis, and multi-modal approaches that use multiple data sources. The study also investigates the challenges of implementing these technologies, such as data privacy concerns, real-time applicability, and individual variability in stress responses. The findings highlight the ability of AI-driven stress detection models to improve mental health monitoring and provide timely interventions to students.