TY - JOUR AU - Sarkar, Angshuman AU - Mallik, Shambo Saurav PY - 2022/10/01 Y2 - 2024/03/29 TI - A Survey of Machine Learning Based Methods for the Diagnosis of Mental Health JF - Asia-Pacific Journal of Management and Technology (AJMT) JA - AJMT VL - 3 IS - 2 SE - DO - 10.46977/apjmt.2022.v03i02.004 UR - https://ejournal.lincolnrpl.org/index.php/ajmt/article/view/85 SP - 39-55 AB - <p><strong>Introduction:</strong> Mental illnesses like depression, schizophrenia, bipolar disorder, etc. have become widespread in today’s society. More than 7.5% of Indians suffer from some kind of mental disorder. Early detection of mental illness is important for treatment as well as to prevent self-harm. Traditionally, the diagnosis involved answering a specifically designed questionnaire at the doctor’s clinic. <strong>Methods</strong>: Mental health data, however, can also be collected from other sources like social media posts, wearable smart-devices, etc. Manual analysis of the patient data may not reveal all the information. Hence, diagnostic errors are common. In recent times Machine Learning (ML) algorithms have been successfully employed for the identification of critical symptoms, the development of diagnostic modules, and the personalization of therapy. <strong>Result</strong> : Authors have systematically reviewed 40 papers that used Machine Learning based techniques for the diagnosis of mental disorders and chose 25 of them for the survey. This paper provides an overview of the application of ML in mental healthcare. It has been focused here on the state-of-the-art Machine Learning based work on mental health. <strong>Discussion:</strong> The majority of the papers consulted so far concentrated on finding whether a subject belongs to a diagnostic category. Classifying into one broad category, however, does not take into account the variability of the symptoms. <strong>Conclusion:</strong> The authors identified some major research prospects in the early detection of schizophrenia.</p> ER -