Machine Learning Based Evaluations of Stress, Depression, and Anxiety

Abstract

One of the most severe issues in modern mental health care is depression, which affects people of all ages and sexes equally. Nowadays, people do not have to work as hard physically and have more stress overall, all of which can lead to mental disruption. Depression results from several factors, including workplace culture, daily life pressures, difficulty maintaining healthy relationships, and the monotony of social and family obligations. In addition, angina and heart attacks have both been linked to depression. Depression is a medical condition that significantly affects how people feel and behave daily. Sadness, anxiety, a lack of interest in activities and work, and, in extreme cases, suicidal thoughts result from this mental condition. This research analyses and compares several Machine Learning Algorithms' outcomes for their ability to predict depressive episodes based on the parameters used. Neural network algorithms generate 97.2% classification accuracy for depression prediction.