A Hybrid Algorithm Combining an Evolutionary Algorithm and a Simulated Annealing Algorithm to Solve a Collaborative Learning Team Building Problem

VIRGINIA YANNIBELLI; ANALÍA AMANDI

LECTURE NOTES IN COMPUTER SCIENCE

Springer

Lugar: Heidelberg; Año: 2013 vol. 8073 p. 376 - 389

0302-9743

In this paper, we address a collaborative learning team building problem that considers a grouping criterion successfully analyzed in the context of software engineering courses. This criterion is based on taking into account the team roles of the students and on building well-balanced teams according to the team roles of their members. To solve the problem, we propose a hybrid algorithm. This algorithm incorporates a simulated annealing algorithm into the framework of an evolutionary algorithm with the aim of improving the performance of the evolutionary search. The simulated annealing algorithm adapts its behavior according to the evolutionary search state. The performance of the hybrid algorithm on ten data sets is compared with those of the algorithms previously proposed in the literature for solving the addressed problem. The obtained results show that the hybrid algorithm significantly outperforms the previous algorithms.