SummarizeLaw

SummarizeLaw is a legal NLP project focused on automatic summarization of Indian court judgments using InLegalBERT. The repository contains sample judgment documents, reference summaries, and a working notebook used to generate and evaluate summaries for legal case texts.

This project is also part of the published research work:

"Automating Court Judgement Prediction and Explanation in Indian Legal Cases"
Springer Link: https://link.springer.com/chapter/10.1007/978-3-032-12827-0_7

---

What this repository includes

• Judgement source files (PDFs and text extracts)
• Reference summaries for evaluation
• Notebook workflow for summarization and experimentation:
  • Summarization_Capstone_ (2).ipynb

---

Method Overview

1. Input legal judgment documents.
2. Apply InLegalBERT-based summarization workflow.
3. Compare generated summaries with reference summaries.
4. Track evaluation performance (reported metric: 86.8667% in this project context).

---

Repository Structure

.
├── case files/               # Input legal case PDFs
├── Judgements_folder/        # Text judgments
├── Reference_Summaries/      # Reference summaries used for evaluation
├── Summarization_Capstone_ (2).ipynb
└── README.md

---

Notes

• This repository is research-oriented and intended for experimentation/learning in legal document summarization.
• You can extend it by:
  • adding reproducible training/inference scripts,
  • packaging the notebook logic into modules,
  • and publishing benchmark details (dataset split, metric definitions, and baseline models).