⚖️ LegalMind AI — Contract Analyzer

LegalMind AI is an intelligent contract summarization tool for legal professionals. It uses cutting-edge NLP models to generate concise summaries of long legal documents, helping save time and improve comprehension.

---

🚀 Features

• 🔍 AI-Powered Summarization using Transformer models (e.g. facebook/bart-large-cnn)
• ⚡ GPU Acceleration if available, for faster performance
• 📥 Input via file upload, direct text, or sample contract
• 📊 Dynamic summary length tuning for optimal output
• 🧠 Summary metrics: original length, compression ratio, processing time
• 🧾 Analysis history with detailed summaries and full document view
• 💾 SQLite database with content hashing to prevent duplicates
• 🎨 Sleek Streamlit UI with custom CSS and tabbed layout

---

🛠️ Setup Instructions

1. Clone the repository

git clone https://github.com/BaraaHazzaa/legal-analyzer.git
cd legal-analyzer

2. Install dependencies

Make sure Python 3.8+ is installed.

pip install -r requirements.txt

If requirements.txt is missing, create it manually with:

streamlit
transformers
torch

3. Run the app

streamlit run app.py

---

💡 Sample Usage

• Upload a .txt contract file, or paste text manually
• Click "Analyze Contract"
• View the AI-generated summary and metrics
• Browse past analyses in the History & Insights tab

---

📁 Project Structure

legal-analyzer/
├── app.py                  # Streamlit frontend + app logic
├── model.py                # Summarization model loading
├── .gitignore
├── contract_analytics.db   # (Ignored) SQLite DB for history
├── model_cache/            # (Ignored) Local HuggingFace cache
├── test_contract.txt       # Sample contract text
└── README.md               # You're here!

---

📌 Notes

• Uses GPU if available (torch.cuda.is_available())
• Local caching for models speeds up load time
• Summaries stored in SQLite with content hashing

---

✅ TODO / Future Improvements

• PDF and DOCX parsing support
• User authentication & per-user history
• Export summaries as PDF/Word
• Search/filter saved analyses

---

👤 Author

Baraa Hazzaa
GitHub