Legal Area Classification: A Comparative Study of Text Classifiers on Singapore Supreme Court Judgments

TL;DR

This study compares various machine learning methods for classifying Singapore Supreme Court judgments into legal areas, highlighting the performance of NLP techniques on a novel, lengthy legal dataset.

Contribution

It introduces a new dataset of Singapore legal judgments and evaluates the effectiveness of different ML and NLP classifiers in legal text classification.

Findings

All models performed well with limited data

State-of-the-art NLP methods show promise but need further optimization for legal texts

Traditional models remain competitive in legal classification tasks

Abstract

This paper conducts a comparative study on the performance of various machine learning (``ML'') approaches for classifying judgments into legal areas. Using a novel dataset of 6,227 Singapore Supreme Court judgments, we investigate how state-of-the-art NLP methods compare against traditional statistical models when applied to a legal corpus that comprised few but lengthy documents. All approaches tested, including topic model, word embedding, and language model-based classifiers, performed well with as little as a few hundred judgments. However, more work needs to be done to optimize state-of-the-art methods for the legal domain.