Topological mapping grounded in particle swarm optimization with evolutionary computation

Abstract

This paper presents a model of topological mapping, which can be realized to compress huge amounts of information and to find their trends and relations. The model is based on particle swarm optimization with evolutionary computation that is a hybrid in concepts of evolutionary algorithm and particle swarm optimization. Particle swarm optimization is an algorithm which we can refer to the collective motion such as birds or fishes, and it is an algorithm that we can find the best solution by using multiple particles. Particle swarm optimization is so exible that we can adapt it to a number of optimization problems. The result of topological mapping can be regarded as one of the optimization problems in the sense to obtain the best positions for mapped data. In order to show the effectiveness of the proposed model, we have examined the abilities for topological mapping by testing input data. © 2012 ICIC International.