Analysis of Machine Learning Algorithms for the Detection of Mental Health Problems in Employees

Abstract

Mental health-related issues are more critical issues in the workplace. Mental health has been a major and challenging issue, especially for working professionals. A bad work environment can cause a variety of physical and physiological problems as well as decreased productivity. People are reluctant to ask for help from mental health specialists, according to studies. The fundamental reason for this is the stigma attached to mental illness. Working professionals therefore have a higher risk of mental health issues. To analyze the consequences of mental health in professionals, in this study, we have analyzed different factors related to personal, professional, and family history using different machine learning algorithms. The algorithms used for the analysis of mental health problems are Naïve Bayes, Logistic Regression, Adaptive Boosting (AdaBoost), Random Forest, and Decision Table. The result obtained from the model training shows that the best performance was achieved using the Random Forest algorithm with an accuracy score of 83.3%. In terms of precision, the AdaBoost algorithm performed best with a precision score of 84.6%. Mental health issues at the workplace are a critical issue and it is believed that the findings from this study will help to aid in medical health to minimize mental health issues at the workplace.