Student Stress Patterns: A Data Driven Exploration using Machine Learning Amidst Academic Challenges and Health Impacts

Student performance is most often hampered by mental health difficulties. Students' motivation, attention, and social ties can all be impacted by mental illness, all of which are key factors in their academic achievement. Despite the widespread use of emergency remote learning (ERL) by educational institutes during the COVID-19 pandemic, little is known about the elements that influence student satisfaction and stress levels in this innovative learning environment in a crisis. Our research uses machine learning algorithms to predict the stress faced by students based on their academic routines, thus providing a timely assessment of the pandemic's impact on students' stress levels. Data collected through student surveys relating to factors such as time spent on studying, social media, health and fitness etc. provide a strong basis to determine students' stress levels. Via supervised machine learning algorithms such as Naive Bayes, Random Forest, Artificial Neural Networks (ANN) etc., predictions are made on academic stress levels by analyzing the prime factors affecting the issue at hand, and a comprehensive comparison is performed with the proposal of the most optimum algorithm for the prediction of stress level.