Modelagem matemática para recomendação de orientadores em programas de pós-graduação stricto sensu: uma abordagem baseada na análise de métricas acadêmicas e científicas

Abstract

The selection of academic advisors and stricto sensu graduate programs is an increasingly complex and crucial process for students, requiring the analysis of a vast amount of data related to programs, research groups, research lines, as well as factors such as academic compatibility, advisor reputation, scientific production, and collaboration networks. Traditionally, this decision is made manually and often based on subjective perceptions and informal advice, which can lead to poorly grounded choices misaligned with students’ academic and professional goals. To support this challenge, this project proposes the development of a mathematical and computational model that integrates data science techniques to calculate a Recommendation Index (Índice de Recomendação (IR)) based on academic indicators extracted from Lattes Platform curricula and other international databases. This model aims to provide objective recommendations, considering criteria such as scientific productivity, mentoring experience, and research area alignment. The model validation was conducted through statistical analyses applied to faculty curricula extracted from the Lattes Platform, including data completeness tests, descriptive analysis of metrics, identification of outliers, and evaluation of the consistency and robustness of the results. Thus, the proposed approach seeks to foster a more objective, well-founded, and individualized process for selecting academic advisors and graduate programs.