Published in: Journal of Computer Science and Engineering in Innovations and Research (ISSN No: 3049-1762 online)
Publication Date: May 15, 2025
The discovery of stress among university students has come to be a common area of study, particularly due to its long-term implications on cognition and mental health. In this present comprehensive review, we adopt a systematic approach to describe the recent trends in a variety of methods employed for detecting stress.
The methods employed are machine learning (ML) and deep learning (DL) methods and physiological parameters as secondary markers of stress. We present the most prominent findings, recognize the existing limitations, and specify the areas of research that need further investigation, based on the information from 20 recent studies in the area.
In addition, the review also identifies key limitations in the existing body of literature, most importantly, the requirement for the development of personalized models, integration of real-time monitoring functions, and fusion of heterogeneous data sources, which would be preferable for improving our understanding and regulation of stress.
Sentiment Analysis, Stress Detection, Machine Learning, Deep Learning, Physiological Data, Academic Stress, AI-Based Stress Management