Predicting the Law Area and Decisions of French Supreme Court Cases

TL;DR

This study applies text classification techniques to predict French Supreme Court case outcomes, law areas, and ruling times, demonstrating high accuracy and exploring factors like temporal influence and judge motivation masking.

Contribution

It introduces a novel application of text classification to legal case prediction, including temporal analysis and motivation masking considerations.

Findings

96% F1 score in predicting case rulings

90% F1 score in predicting law areas

75.9% F1 score in estimating ruling time span

Abstract

In this paper, we investigate the application of text classification methods to predict the law area and the decision of cases judged by the French Supreme Court. We also investigate the influence of the time period in which a ruling was made over the textual form of the case description and the extent to which it is necessary to mask the judge's motivation for a ruling to emulate a real-world test scenario. We report results of 96% f1 score in predicting a case ruling, 90% f1 score in predicting the law area of a case, and 75.9% f1 score in estimating the time span when a ruling has been issued using a linear Support Vector Machine (SVM) classifier trained on lexical features.