|
--- |
|
license: apache-2.0 |
|
--- |
|
# 🧠 Mental Health Text Classifier |
|
|
|
This repository contains an AI-powered tool designed to classify text into one of the following mental health categories: |
|
- **Stress** |
|
- **Anxiety** |
|
- **Depression** |
|
- **Other** |
|
|
|
It leverages **OpenAI's GPT-4** for classification using few-shot learning, combined with BM25 for contextual similarity. |
|
|
|
## Features |
|
- **User-Friendly Interface:** Easy-to-use Gradio-powered web application. |
|
- **Interactive Classification:** Real-time text input and predictions. |
|
- **Extensibility:** Reusable `MentalHealthClassifier` model for integration into other projects. |
|
|
|
## How It Works |
|
1. The user provides a text input (e.g., thoughts or feelings). |
|
2. The tool uses BM25 to retrieve similar examples from the training data. |
|
3. GPT-4 analyzes the input and predicts the best-fitting mental health category. |
|
|
|
## Prerequisites |
|
- **Python 3.8+** |
|
- OpenAI API Key (Get one from [OpenAI](https://platform.openai.com/signup/)) |
|
- Hugging Face Token (Optional, for dataset access) |
|
|
|
## Installation |
|
1. Clone the repository: |
|
```bash |
|
git clone https://github.com/your-username/mental-health-classifier.git |
|
cd mental-health-classifier |
|
|
