Using natural language processing methods to predict judicial outcomes

• Autor(a) principal: Bertalan, Vithor Gomes Ferreira
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://www.teses.usp.br/teses/disponiveis/59/59143/tde-04012021-232455/
• topic: Classificador jurídico; Legal classifier; Legal prediction; Natural language processing; Predição jurídica; Processamento de linguagem natural

Natural Language Processing (NLP) and Artificial Intelligence (AI) for the field of Law is a growing area, with the potential of radically changing the daily routine of legal professionals. The amount of text generated by those professionals is outstanding, and to this point, it is a knowledge area to be more explored by Computer Science. One of the most acclaimed fields for the combined area of NLP, AI, and Law is Legal Prediction, in which intelligent systems try to predict specific judicial characteristics, such as the judicial outcome or the judicial class or a given case. This research creates classifiers to predict judicial outcomes in the Brazilian legal system. For this purpose, we developed a text crawler to extract data from the official Brazilian electronic legal systems. Afterward, we developed a dataset of Second Degree Murder and Active Corruption cases, and different classifiers, such as Support Vector Machines and Neural Networks, were used to predict judicial outcomes by analyzing textual features. As a final goal, we used the findings of one of the algorithms, Hierarchical Attention Networks, to find a sample of the most important words used to absolve or convict defendants.