Virtual Self Care Companion – Detection of mental illness using machine learning and deep learning

Main Article Content

Ms. Vaishnavi Iyer, Ms. Tejal Kadlag, Mr. Mayur Patil, Mr. Ashwin Pillai, Prof. K P Moholkar


Smartphones are becoming increasingly ubiquitous. The app market is growing in all the fields, including health and medical consultancy. Although mental health care sector has been slow to adopt technology, the availability and use of health care apps has exponentially grown in recent years. The use of mobile apps to treat mental illness is widespread and growing. The main motive of this paper is to present a survey on the existing chatbot applications as well as study the  different techniques applied to understand the behavioral change and     mood swings of individuals and detect whether or not they have any kind of mental illness .Based on the in-depth survey conducted across 8 different chatbot applications with regards to the features, accessibility, availability and technical specifications ,the research shows that around 70% users were not satisfied with the services provided.  Using multiple categorization models, this paper also seeks to improve the accuracy of predictions (such as Logistic regression, Random Forest classifier, decision tree classifier, Stacking, KNN etc.). This research study is aimed toassist the generality, nature and reasons of mental illness at the primary levels of the conditions and the identification of these disorders. Based on data sources, machine learning techniques, and feature extraction methods, the paper presents a critical assessment analysis on mental health detection.

Article Details