Dissertation defense (December 11, 2024): Fabio da Silva Gregório
Student: Fabio da Silva Gregório
Title: Guided LexRank for Advanced Retrieval in Legal Analysis
Advisor: Eduardo Bezerra da Silva
Committee: Eduardo Bezerra da Silva (Cefet/RJ – PPCIC), Altigran Soares da Silva (IComp / UFAM), Kele Teixeira Belloze (PPCIC / CEFET-RJ), Gustavo Paiva Guedes e Silva (PPCIC / CEFET-RJ).
Day/Time: December 11, 2024 / 4 p.m
Abstract: The Brazilian Constitution provides mechanisms for citizens to petition the Judiciary, including the so-called special appeal. This specific type of appeal aims to standardize the legal interpretation of Brazilian legislation. The processing of special appeals is one of the daily tasks of the Judiciary, regularly presenting significant demands in its courts. We propose a method, based on unsupervised machine learning, to assist the legal analyst in classifying a special appeal on a topic from a list made available by the National Court of Brazil (STJ). As part of this method, we propose a modification of the graph-based LexRank algorithm, which we call Guided LexRank. This algorithm generates the summary of a special appeal. The degree of similarity between the generated summary and different topics is evaluated using the BM25 algorithm. As a result, the method presents a ranking of themes most appropriate to the special appeal analyzed. The proposed method does not require prior labeling of the text to be evaluated and eliminates the need for large volumes of data to train a model, as is typically the case in supervised models. We evaluate the effectiveness of the method by applying it to a special appeal corpus previously classified by human experts.
