Legal Text Named Entity Recognition (NER) Project

Project Overview

This independent project was dedicated to the development of a Named Entity Recognition (NER) system tailored for legal texts. Utilizing the PyTorch framework, the project implemented a Bi-directional Long Short-Term Memory (BiLSTM) network combined with a Conditional Random Field (CRF) layer to effectively identify and classify named entities within legal documents.

Achievements

• High Accuracy: Successfully achieved 88% accuracy on the test set, demonstrating the effectiveness of the BiLSTM + CRF architecture in the context of legal text NER.
• Advanced Model Architecture: The use of BiLSTM allowed for capturing contextual information from both past and future tokens, while the CRF layer modeled the dependencies between labels for accurate entity classification.

Repository

For more details on the project’s methodology, implementation, and results, visit the GitHub repository: Legal Text NER Project Repository.