Sentiment Analysis for Therapy Chatbots: A Comparison of Supervised Learning Approaches
Bhawani Singh Rathore, Sandeep Chaurasia · 2024
The maj ority of people in the world presently suffer from mental illness, and many of them are not even aware of it. Mental health awareness is important, but fear and ignorance often prevent people from talking about their issues. This work investigates the incorporation of sentiment analysis into therapy chatbots by utilizing deep learning and machine learning techniques, acknowledging the significance of mental health. By enabling users to interact with chatbots in a conversational manner, these systems aim to revolutionize mental health support. We evaluate various supervised learning algorithms for sentiment analysis within therapy chatbots, emphasizing their practicality and effectiveness. Our study reveals the Random Forest (RF) algorithm as a standout performer, achieving an impressive accuracy rate of 87%. This comparative analysis offers insights into the potential of these algorithms to enhance mental health assistance and user engagement, contributing to the advancement of conversational AI in mental health care. In light of our findings, this research underscores the significance of employing sentiment analysis within therapy chatbots and provides valuable insights into the comparative performance of supervised learning approaches, aligning with the overarching goal of enhancing mental health support through innovative technological solutions.