MindMatch is a secure, AI-powered mental health support platform designed to triage and monitor user messages in real time. It uses natural language processing (NLP) to detect emotional distress and assign severity scores to incoming messages. Counsellors and support staff can view alerts and prioritize users who may be at risk, ensuring faster, more focused intervention.
MindMatch runs on a hybrid backend architecture using FastAPI, SQLAlchemy and HuggingFace models. The frontend is built in React, and all data is handled with anonymisation.
MindMatch was developed to support mental health organizations, universities, and helpline networks by automating the first stage of triage in digital chat systems.
The platform analyses text messages in real-time, assigns a distress severity score (0–1), and triggers alerts when necessary. This reduces response time for high-risk users and allows counsellors to focus where help is most needed.
- 🧠 Real-time NLP Distress Detection – Detects sadness, fear, and suicidal ideation from text.
- 🔔 Automated Alerts – Triggers SMS notifications when risk thresholds are exceeded.
- 🌐 Anonymous Chat Support – Protects user identity while enabling meaningful outreach.
- 🧩 Scalable Backend – Built using FastAPI and SQLAlchemy with modular architecture.
- MindMatch is available under the MIT License.