A Proposal for Virtual Mental Health Assistant
The proposed assistant will have a chat feature, psychological assessment, an emotion detection module and a recommendation system for improving the mood of the user, and has shown comparatively higher accuracy for Naive Bayesian model.
Abstract
In these current times, mental health has become one of the most neglected and yet one of the most serious aspects of our overall well being. This paper proposes a system for a virtual mental health assistant owing to financial, time and space constraints and shortage of resources related to in-person therapy. Of tentimes, disturbed mental health is a snowball effect built up over time and requires continuous attention and conscious efforts to improve. This is possible with the help of a virtual mental health assistant. The proposed assistant will have a chat feature, psychological assessment, an emotion detection module and a recommendation system for improving the mood of the user. We have used Naive Bayes classifier and Neural Networks for sentiment analysis. Our system has shown comparatively higher accuracy for Naive Bayesian model.