Legal documents are complex and unstructured, making manual extraction of important entities inefficient and error-prone. This project presents an automated Named Entity Recognition (NER) system for legal documents using a fine-tuned Legal-BERT model. The system identifies key entities such as persons, organizations, dates, locations, and legal provisions. A Streamlit-based web application enables users to upload documents and view extracted entities interactively. The proposed solution reduces manual effort and improves the efficiency of legal document analysis.
S et al. (Wed,) studied this question.