A utilização de Algoritmos Genéticos para a formação de equipes heterogêneas em turmas universitárias

Abstract

The traditional way of learning through teams is a historical aspect pertinent to Education. However, the difficulty in distributing students in a way that favors learning among individuals is still a problem with this approach. On the other hand, technological and computational advancement has enabled the creation of methods and techniques capable of identifying potential results to the problems in which they are humanly impossible to solve. Through the use of metaheuristics, it is possible to explore the search space in order to obtain a better configuration of the factors that approach the optimal state. In this context, the present work proposes an algorithmic solution based on Genetic Algorithms to optimize the process of team formation in competences domain, favoring collaborative learning through the exchange of knowledge among individuals. The experiments realized that the proposed strategy was able to maximize the heterogeneity in the teams, unlike the traditional methods commonly used for clustering. Furthermore, through the analysis of variance, it was verified that, in fact, the proposed algorithm is a potential technique for guaranteeing solutions that are capable of grouping students with heterogeneous characteristics, thus overcoming the methods still employed for this purpose.