Towards Intelligent Legal Document Analysis: CNN-Driven Classification of Case Law Texts

TL;DR

This paper introduces a lightweight CNN-based framework for legal document classification that outperforms larger models in accuracy and speed, demonstrating efficiency and scalability in legal AI applications.

Contribution

The work presents a novel CNN-driven classification system with lemmatisation and subword embeddings, achieving high accuracy and efficiency in legal document analysis.

Findings

Achieved 97.26% accuracy and 96.82% macro F1-score on legal texts.

Model operates with only 5.1 million parameters and 0.31 ms inference time per document.

Outperforms BERT and other baselines in accuracy and speed.

Abstract

Legal practitioners and judicial institutions face an ever-growing volume of case-law documents characterised by formalised language, lengthy sentence structures, and highly specialised terminology, making manual triage both time-consuming and error-prone. This work presents a lightweight yet high-accuracy framework for citation-treatment classification that pairs lemmatisation-based preprocessing with subword-aware FastText embeddings and a multi-kernel one-dimensional Convolutional Neural Network (CNN). Evaluated on a publicly available corpus of 25,000 annotated legal documents with a 75/25 training-test partition, the proposed system achieves 97.26% classification accuracy and a macro F1-score of 96.82%, surpassing established baselines including fine-tuned BERT, Long Short-Term Memory (LSTM) with FastText, CNN with random embeddings, and a Term Frequency-Inverse Document Frequency (TF-IDF) k-Nearest Neighbour (KNN) classifier. The model also attains the highest Area Under the Receiver Operating Characteristic (AUC-ROC) curve of 97.83% among all compared systems while operating with only 5.1 million parameters and an inference latency of 0.31 ms per document - more than 13 times faster than BERT. Ablation experiments confirm the individual contribution of each pipeline component, and the confusion matrix reveals that residual errors are confined to semantically adjacent citation categories. These findings indicate that carefully designed convolutional architectures represent a scalable, resource-efficient alternative to heavyweight transformers for intelligent legal document analysis.