Date created: | Last Updated:

Category: Project

Description: The Brazilian judiciary faces a significant workload, leading to prolonged durations for legal proceedings.In response, the Brazilian National Council of Justice introduced the Resolution 469/2022, which provides formalguidelines for document and process digitalization, thereby creating the opportunity to implement automatic tech-niques in the legal field. These techniques aim to assist with various tasks, especially managing the large volume oftexts involved in law procedures. Notably, Artificial Intelligence (AI) techniques open room to process and extractvaluable information from textual data, which could significantly expedite the process. However, one of the chal-lenges lies in the scarcity of datasets specific to the legal domain required for various AI techniques. Obtaining suchdatasets is difficult as they require some expertise for labeling. To address this challenge, this article presents fourdatasets from the legal domain: two include unlabelled documents and metadata, while the other two are labeledusing a heuristic approach designed for use in textual semantic similarity tasks. Additionally, the article presents asmall ground truth dataset generated from domain expert annotations to evaluate the effectiveness of the proposedheuristic labeling process. The analysis of the ground truth labels highlights that conducting semantic analysisof domain-specific texts can be challenging, even for domain experts. Nonetheless, the comparison between theground truth and heuristic labels demonstrates the utility and effectiveness of the heuristic labeling approach.