🚀 Features

• 🔍 Real-time hate speech detection
• 🌐 Interactive React-based frontend
• 🧠 ML model hosted on Hugging Face Hub
• 🐍 Flask API for backend inference
• 📊 Confidence score for predictions

🧠 Model

The detection model is trained on Hate Speech and Offensive Language Dataset and fine-tuned for binary/multiclass classification (e.g., hate, offensive, neither). Hosted publicly on Hugging Face:

• Model Name: KhalidGB/HateBERT
• Model Card: View on Hugging Face

🖼️ Demo

Try the model directly on Hugging Face 🤗

hate-speech-detector/
│
├── venv/       
├── model/        # Fine-tuned hugging face model
│   └── config.json
    └── model.safetensors
    └── special_tokens_map.json
    └── tokenizer_config.json
    └── training_args.bin
    └── vocab.txt
├── README.md
├── app.py
└── requirements.txt
└── LICENSE
└── README.md

make sure you download The HateBERT model from Hugging Face into model folder

⚙️ Installation

Backend (Flask)

python -m venv venv
source venv/bin/activate #Mac
venv\Scripts\activate    #Windows
pip install -r requirements.txt
flask run

Frontend (https://github.com/khalid21456/Hate-Speech-Detector-UI)

cd frontend
npm install
npm run dev

Make sure the backend is running on http://localhost:5000 and frontend on http://localhost:8080.