Abstract
Mental health disorders are one of the most significant public health problems worldwide. Currently, one in every eight individuals is suffering from some kind of mental health issue. Anxiety and depression are common manifestations of neurotic and psychotic problems respectively. Screening and diagnosis of mental health disorders in the population is a complex clinical task. The conventional approach of questionnaire-based interviews and detailed medical examination for screening and diagnosis of anxiety and depression requires highly trained healthcare professionals like psychiatrists or psychologists and a significant amount of time and patience. Machine learning (ML) is state-of-the-art technology where computers can learn to perform a task from the data. So, the task of screening mental health disorders can also be performed using ML algorithms. Attributes like job profile, age, marital status, employment status, duration of service, working hours, and chronic disease conditions were selected to predict mental health disorders. For this research work three popular classification algorithms i.e., decision tree, logistic regression, and random forest were selected based on the literature review and evaluated based on Accuracy, precision, recall, and f1 score. A random forest model with hyperparameter tuning was found to be the best fit for this specific purpose.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call