verBERT: Automating Brazilian Case Law Document Multi-label Categorization Using BERT

TL;DR

This paper explores the use of BERT-based models to automate the multi-label categorization of Brazilian case law documents, achieving significant performance improvements over baseline methods.

Contribution

It introduces a multi-label BERT approach tailored for Brazilian legal documents, demonstrating its effectiveness with substantial F1-score gains.

Findings

Achieved micro-averaged F1-Score of 0.72

Gained 30 percentage points over baseline

Validated the approach on datasets from the Kollemata Project

Abstract

In this work, we carried out a study about the use of attention-based algorithms to automate the categorization of Brazilian case law documents. We used data from the Kollemata Project to produce two distinct datasets with adequate class systems. Then, we implemented a multi-class and multi-label version of BERT and fine-tuned different BERT models with the produced datasets. We evaluated several metrics, adopting the micro-averaged F1-Score as our main metric for which we obtained a performance value of F1-micro=0.72 corresponding to gains of 30 percent points over the tested statistical baseline. In this work, we carried out a study about the use of attention-based algorithms to automate the categorization of Brazilian case law documents. We used data from the \textit{Kollemata} Project to produce two distinct datasets with adequate class systems. Then, we implemented a multi-class and multi-label version of BERT and fine-tuned different BERT models with the produced datasets. We evaluated several metrics, adopting the micro-averaged F1-Score as our main metric for which we obtained a performance value of $\langle \mathcal{F}_1 \rangle_{micro}=0.72$ corresponding to gains of 30 percent points over the tested statistical baseline.