A Comprehensive Model for Cognitive Stress Detection for College Students

For people in today’s world, stress is a prevalent issue that affects an individual’s health, productivity, and well-being. This was majorly seen among the university students as stress modifies their behaviour, ideas and feelings which impacts overall academic performance and mental health. As this is a common issue, there is a growing interest in developing models for stress detection to ensure timely resolution. The detection of stress is quite challenging, as analysis of different sources of data including physiological signals, facial expressions, daily activities, and social interactions is involved. This paper provides a comprehensive study of current machine learning-based stress detection methodologies based on data acquisition, feature extraction, and classification algorithms. This work demonstrates detection of stress through training of machine learning model on the labelled datasets along with identification of stress from the facial expressions. Results shows the efficient accuracy rates indicating potential for timely resolution and real time monitoring by various means. This paper adds to the technological solutions used for managing stress levels and gives insights into the effective approaches towards enhancing mental health of university students.