H. Grass Boada, J. López Fidalgo, E. Benitez, C. De La Calle Arroyo
In the past decade, there has been an increasing interest in the analysis of complex networks due to their applications in several contexts. In the context of large-scale optimization problems, a innovate class of algorithms developed, the so-called metaheuristics. One of the issues to attend in the metaheuristic algorithms is the need of setting the values of several components and parameters within them.
Design of Experiments is well established theoretically and offers efficiency in terms of the amount of data that needs to be gathered, which is critical when attempting to understand immense algorithm design spaces. Therefore, with the aim to obtain the best performance of the metaheuristic algorithms, a systematic method of tuning up these parameters should be developed. In our case, to analyze the parameter tuning approach in metaheuristics, a Genetic Algorithm is used due to its wide application area, including community detection in complex networks.
Palabras clave: Community Detection, metaheuristics, Response surface methodology
Programado
GT06.DEX1 Sesión Invitada
7 de noviembre de 2023 18:40
CC4: Sala 2