ifkad articles

Brazil's University Ranking: a Knowledge Prediction Study with Machine Learning

Sérgio Nicolau Silva, Cleverson Tabajara Vianna, Fernando Alvaro Ostuni Gauthier, Antônio Pereira Cândido

How to distinguish the best or worst institutions of higher education? This is a question that permeates the minds and hearts of parents, students, and teachers because education is an investment in the personal and nation’s future. As a source of information for the response to asking, the University Ranking of Folha – RUF appears. Known for its traditional evaluation, the Folha’s Ranking is considered an independent evaluation tool and provides a ranking of the best Brazilian universities. 74% of the data are related to research areas and postgraduate programs. Who regulates and supervises the postgraduate programs in Brazil is CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), authorizing or not the program, assigning a score from 1 to 7, with 7 being the best score. Your data for this evaluation is published. In this article, are using machine learning techniques based on Naïve Bayes algorithms. CAPES data and the Folha’s Ranking of previous years are used as the training mass for the machine Naïve Bayes algorithm. After the training, CAPES data from 2015 was applied to predict the 2016 Ranking with a hit rate of 61.5%. A percentage above 60 of the Folha’s Ranking shows that it is possible, with a more detailed study and analysis of the techniques, to predict with a certain confidence. It should be noted that according to the Folha’s Ranking roles, the Scientific Research (mostly postgraduate) corresponds to a weight of 42% in the ranking.

IN: Proceedings IFKAD 2018 – Societal Impact of Knowledge and Design
PP: 609-620