Author Affiliations
[1] [2] [3] [4] Dept. of Computer Science and Engineering, Er. Perumal Manimekalai College of Engineering, Hosur, Tamilnadu, India.
[5] Associate professor, Department of Computer Science and Engineering, Er. Perumal Manimekalai College of Engineering, Hosur.
[6] Assistant Professor, Department of Computer Science and Engineering, Er. Perumal Manimekalai College of Engineering, Hosur.
Abstract
Mental health remains a critical public health challenge globally, with millions of individuals lacking access to timely professional support. This paper presents MindBridge, an AI-powered anonymous peer mental health support platform that enables real-time, stigma-free conversations among users while continuously monitoring message content for crisis signals. The system employs a four-layer Natural Language Processing (NLP) moderation pipeline integrating keyword-based crisis detection, Detoxify machine learning toxicity classification, a RoBERTa-based sentiment analysis model fine-tuned on social media text, and a weighted riskscoring algorithm. The backend is built on Django REST Framework with Django Channels for persistent WebSocket communication, while the frontend is developed using React.js. Messages are classified into five sentiment states and four risk levels, with high-risk events automatically generating crisis alerts visible to administrators. Five thematic chat rooms covering anxiety, depression, stress, loneliness, and general support are provided. Experimental evaluation demonstrates accurate sentiment classification, sub-second moderation latency, graceful degradation to keyword-only mode when ML models are unavailable, and a complete real-time presencetracking system. MindBridge bridges the gap between peer support communities and AI-assisted safety monitoring.
Keywords: Mental health, anonymous chat, NLP moderation, WebSocket, Django Channels, RoBERTa, sentiment analysis, real-time systems, crisis detection, peer support.
How to Cite This Article
Karthick S, Dharshan V, Jagan Raj M, Abilash K, Dr.N.Shunmuga karpagam, M Keerthika (2026). MindBridge: An AI-Powered Anonymous Peer Mental Health Support Platform with Real-Time NLP Moderation. International Journal of Innovative Research in Multidisciplinary Education & Technology (IJIRMET), 11(3).