Identification of Health Crisis During COVID-19 Using Machine Learning Techniques

Abstract

Psychiatry began to develop empirical approaches only in the past 30 years to conceptualize, assess, and
document positive mental health. In society, we accept mental disorders as normal and try to hide them.
In this article, the authors came across the fact of how severe Mental Illness affects our lives during
COVID-19 and the measures that can be taken to detect the same. Authors have developed methods to
predict an individual's depressive factors. Primary data have been collected from people aged 18 to 59.
Neurotic suffering cannot be fully understood without understanding real health and a sound approach
towards life so, the authors have taken the approach of all the factors including depression, anxiety,
stress, eating disorder, short temper, hallucination[1], loneliness, suicidal attempt, constant guilt feeling,
fearful all these and more, realistic disturbing state and trait that results in various mental diseases have
been considered and analysed. The authors used Gain Ratio with Random Forest, Naïve Bayes, Decision
Tree, and SVM, and Information Gain with Random Forest, Naïve Bayes, Logistic Regression, Decision
Tree, and SVM to predict an individual's depression.