Empath.ai - Mental Health Support Chatbot

A compassionate AI-powered chatbot that provides emotional support and mood tracking capabilities.

🌟 Features

• Empathetic Chat Interface: AI-powered conversations that detect emotional tone and respond with empathy.
• Advanced Emotion Detection: Uses the GoEmotions model to detect 28 distinct emotions (e.g., Joy, Grief, Admiration, Remorse) for highly nuanced responses.
• Crisis Intervention: Built-in safety mechanisms to detect crisis keywords and provide immediate helpline resources.
• Mood Analytics: Track your emotional patterns over time with visual dashboards.
• Coping Strategies: Evidence-based techniques for emotional wellbeing.
• Secure Authentication: User accounts with password hashing and session management.
• Privacy-Focused: All conversations are processed locally.

🏗️ Architecture

Frontend (/frontend)

• Framework: React with TypeScript
• State: React Hooks + Local Storage
• UI Library: shadcn/ui components + Tailwind CSS
• Structure:
  • src/pages: Main views (Login, Chat, Dashboard)
  • src/api-int: Centralized API integration
  • src/components: Reusable UI components

Backend (/backend)

• Framework: Flask with CORS
• Database: SQLite (local mindfulchat.db)
• Auth: Bcrypt password hashing + Server-side sessions
• Endpoints:
  • /auth/*: Login/Register
  • /chat: Main chat interface
  • /mood/*: Mood tracking & analytics

ML Model (/mlmodel)

• Model: SamLowe/roberta-base-go_emotions
• Capabilities: Classifies text into 28 emotional categories.
• Logic: Custom response templates for every emotion.

🚀 Getting Started

Prerequisites

• Node.js (v16+) & npm
• Python 3.8+ & pip

1️⃣ Backend Setup

The backend handles the AI logic and database.

cd backend

# Create virtual environment (Recommended)
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Start the server
python app.py

Server runs at http://localhost:5000

2️⃣ Frontend Setup

The frontend is the React user interface.

cd frontend

# Install dependencies (First time only)
npm install

# Start development server
npm run dev

Frontend runs at http://localhost:5173

📁 Project Structure

HACKATHON/
├── frontend/           # React Application
│   ├── src/
│   │   ├── pages/      # Login, Chat, Dashboard
│   │   ├── api-int/    # API Services
│   │   └── components/ # UI Elements
│   ├── package.json
│   └── vite.config.ts
├── backend/            # Flask Server
│   ├── app.py          # Main Entry Point
│   ├── database.py     # SQLite Manager
│   └── tests/          # Unit Tests
├── mlmodel/            # AI Logic
│   ├── emotion_analyzer.py
│   └── chatbot.py
└── README.md

🧠 AI Model

We utilize the SamLowe/roberta-base-go_emotions model to identify 28 distinct emotions: Admiration, Amusement, Anger, Annoyance, Approval, Caring, Confusion, Curiosity, Desire, Disappointment, Disapproval, Disgust, Embarrassment, Excitement, Fear, Gratitude, Grief, Joy, Love, Nervousness, Optimism, Pride, Realization, Relief, Remorse, Sadness, Surprise, Neutral.

⚠️ Important Notice

This is a supportive tool, not a replacement for professional mental health care. If you are in crisis, the chatbot is designed to provide emergency resources immediately.

🤝 Contributing

This project was created for a hackathon. Contributions are welcome!