This project leverages advanced AI models to summarize, simplify, and analyze legal documents efficiently. It integrates Named Entity Recognition (NER), extractive summarization, and abstractive summarization to ensure high-quality legal text processing.
- Fine-tuned LegalBERT model for legal-specific NER.
- Identifies legal entities such as sections, acts, case references, and NGO policies.
- Helps in extracting key information from government legal documents and NGO policy papers.
- Utilizes BERTSUM to extract the most relevant sentences.
- Works best for lengthy legal documents without altering meaning.
- Enables quick understanding of critical legal arguments and rulings.
- Utilizes legal domain-specific datasets with structured annotations.
- Covers sections such as:
- Legal Sections: (e.g., Article 21, IPC Section 420)
- Acts & Statutes: (e.g., Environmental Protection Act, 1986)
- NGO-Related Laws: (e.g., Foreign Contribution Regulation Act (FCRA))
- Case References: (e.g., Vishaka v. State of Rajasthan, 1997)
- Uses ROUGE (Recall-Oriented Understudy for Gisting Evaluation) for measuring textual overlap.
- Supports BLEU (Bilingual Evaluation Understudy) for fluency and accuracy.
- Additional metrics: BERTScore for semantic accuracy.
- Users can ask context-based questions related to the document.
- AI-powered Q&A system extracts precise answers.
- Extending summarization to regional Indian languages.
- Ensuring accessibility for non-English legal documents.
# Clone the repository
git clone https://github.com/your-repo/legal-summarizer.git
cd legal-summarizer
# Install dependencies
pip install -r requirements.txt
# Run the model
python summarize.py --input legal_document.pdf --output summary.txt- Integration with Spacy for advanced legal NLP.
- Support for summarizing legal contracts & compliance documents.
- Interactive dashboard for legal analysis & visualization.
This README provides an overview of the core functionalities and future plans. Let me know if you want any refinements! 🚀