AI-Driven Chatbot for Mental Health Support Using Sentiment Analysis

V. Mythily, D. Vinoparkavi, P. Sukumar, Chanchalhas V., Ajith M., Barani Kumar K. M.

2025

Abstract

The Mental care remains a problem due to its low accessibility. This paper proposes AI-controlled chatbots based on mood analysis to identify emotions and submit appropriate answers. The system uses NLP and machine learning to provide emotional care in real-time and self-help materials. The system is confidential, scalable and provides early intervention for mental wells. For the majority of the population, psychological health care is still not easy for the surrounding stigma. The proposed paper provides an AI-driven chatbot that recognizes emotions and provides appropriate responses through mood analysis. The system integrates natural language processing and algorithms for machine learning to provide real-time emotional support for self-therapy and provide resources for self-therapy. Created for privacy scalability and affordability the bot offers a subtle, private midfield for anyone who wants to receive emotional support. Then it may be the first sign of already needing it and serves as an early intervention tool that may be useful if it is needed. The fusion of gaps between users and healthcare is an easy first step for those who are unsure whether they are willing to rely on what is ultimately perceived as professional help. The experimental results show that the AI-operated chatbot developed here can perform mood analysis with satisfactory quality and recognize emotional nuances while having empathetic conversations. The system generates corresponding answers that are intended to provide comfort and support, not just capture the detection of emotional signs from user statements. It uses techniques such as relaxation exercises and CBT to focus on aggressive reinforcement. It is based on general principles of mental health problem solving with the aim of strengthening concerns for users. The ability to adapt is important to enable timely and real-time responses based on the user's emotional state. Using adaptive learning characteristics this tool based on emotional variables, allows flexible adaptation to a single user in real time. User interactions develop into useful ones that can contribute to treating emotional problems under untreated conditions. Chatbots cannot replace therapy or specialized care. It is easy to achieve the initial intervention for those looking for support before full treatment with a clinician. Chatbots adapt to user interaction in real time to improve understanding of emotional information and provide an appropriate support. It also serves as the first treatment tool for mental health to help people before specialized treatment. Future work will include developing mood analysis models and supporting multilingual support for more people.

Paper Citation

in Harvard Style

Mythily V., Vinoparkavi D., Sukumar P., V. C., M. A. and M. B. (2025). AI-Driven Chatbot for Mental Health Support Using Sentiment Analysis. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 292-297. DOI: 10.5220/0013881700004919

in Bibtex Style

@conference{icrdicct`2525,
author={V. Mythily and D. Vinoparkavi and P. Sukumar and Chanchalhas V. and Ajith M. and Barani M.},
title={AI-Driven Chatbot for Mental Health Support Using Sentiment Analysis},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={292-297},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013881700004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - AI-Driven Chatbot for Mental Health Support Using Sentiment Analysis
SN - 978-989-758-777-1
AU - Mythily V.
AU - Vinoparkavi D.
AU - Sukumar P.
AU - V. C.
AU - M. A.
AU - M. B.
PY - 2025
SP - 292
EP - 297
DO - 10.5220/0013881700004919
PB - SciTePress