Analyzing Various Machine Learning Classifiers for the efficient Prediction of Student Mental Stress
Abstract
A prevalent societal issue that affects people nowadays is mental stress. Stress is typically felt when one feels that the amount of pressure or demand is more than one's ability to handle it. A person's thoughts, actions, emotions, and interpersonal communication can all be impacted by mental health problems. The major issues that student faces now a days that will suffer their mental health are Depression, Addiction, Anxiety, Eating Disorders, Substance Misuse and Suicidal Intent. Some Students also suffers from a Huge Academic Pressure. It might be from their own mind for gaining more & more in their Academics or might be from Parental Pressure. Accurate analysis and prediction of stress patterns may be possible with the use of machine learning techniques and enabling prompt responses. With an emphasis on the function of machine learning models, the influence of physiological and behavioural characteristics, this paper explores the important facets of mental stress detection. The search was conducted on several databases (IEEE, Scopus, Elsevier, and Web of Science). The topmost objective of the paper is to analyze various algorithms that are used to predict the level of stress among an individual. This Review paper is based on the analysis of various approaches and finally gives the most appealing among all. Random Forest & Gradient Boosting are the best algorithm with topmost accuracy that has been used in various papers and also helps in accurately predicting the level of stress among the individual. Keywords: K Nearest Neighbor, Random Forest, Naive Bayes, Regression, Decision Tree, Support vector machine, Gradient Boosting
Cite this paper
Published in
