Towards a new approach to legal indexing using facets

This paper presents work in progress to design and test a new model for indexing judicial decisions using facets. We have created a prototype database with a sample of 2 500 judicial decisions in the fields of administrative law, labor law and the law of obligations. The decisions are indexed using four facets (Person, Action, Thing, Context) and two classes (Legal category, Remedies and sanctions). We will test the prototype to compare its performance with that of a standard database. It is hoped that our model will support the full or partial automation of legal indexing and provide a structure for AI applications in the legal domain