Empirical Study of Deep Learning for Text Classification in Legal Document Review

TL;DR

This paper compares deep learning, specifically CNN, with traditional SVM algorithms for legal document review, showing CNN's superior performance with larger datasets in predictive coding tasks.

Contribution

It provides empirical evidence that CNN outperforms SVM in legal text classification, highlighting deep learning's potential in legal document review.

Findings

CNN performs better with larger datasets

Deep learning shows promise in legal predictive coding

SVM is less effective with increasing data volume

Abstract

Predictive coding has been widely used in legal matters to find relevant or privileged documents in large sets of electronically stored information. It saves the time and cost significantly. Logistic Regression (LR) and Support Vector Machines (SVM) are two popular machine learning algorithms used in predictive coding. Recently, deep learning received a lot of attentions in many industries. This paper reports our preliminary studies in using deep learning in legal document review. Specifically, we conducted experiments to compare deep learning results with results obtained using a SVM algorithm on the four datasets of real legal matters. Our results showed that CNN performed better with larger volume of training dataset and should be a fit method in the text classification in legal industry.