Collaborative Learning Team Formation Considering Team Roles: An Evolutionary Approach based on Adaptive Crossover, Mutation and Simulated Annealing

VIRGINIA YANNIBELLI; ANALÍA AMANDI

Research in Computing Science

National Polytechnic Institute

Lugar: México; Año: 2018 vol. 147 p. 61 - 74

1870-4069

In this paper, a hybrid evolutionary algorithm is proposed to solve a collaborative learning team formation problem in higher education contexts. This problem involves a grouping criterion evaluated satisfactorily in a great variety of higher education courses as well as training programs. This criterion is based on the team roles of students, and implies forming well-balanced teams respecting the team roles of their members. The hybrid evolutionary algorithm uses adaptive crossover, mutation and simulated annealing processes, in order to improve the performance of the evolutionary search. These processes adapt their behavior regarding the state of the evolutionary search. The performance of the hybrid evolutionary algorithm is exhaustively evaluated on data sets with very different complexity levels, and after that, is compared with those of the algorithms previously reported in the literature to solve the addressed problem. The results obtained from the performance comparison indicate that the hybrid evolutionary algorithm significantly outperforms the algorithms previously reported, in both effectiveness and efficiency.