The proposal of a velocity memoryless clustering swarm
Tipo
Artigo de evento
Data de publicação
2010
Periódico
2010 IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 IEEE Congress on Evolutionary Computation, CEC 2010
Citações (Scopus)
13
Autores
Szabo A.
Prior A.K.F.
De Castro L.N.
Prior A.K.F.
De Castro L.N.
Orientador
Título da Revista
ISSN da Revista
Título de Volume
Membros da banca
Programa
Resumo
The PSC (Particle Swarm Clustering) algorithm is an adaptation of the PSO (Particle Swarm Optimization) algorithm, and, therefore, follows a heuristic inspired by the optimization version. The particles move in the search space in order to become representatives of the natural groups of the database. The movement of particles is based on the behavior of social animals, like a flock of birds or a school of fish, which adjust their movements to defend the group and retrieve food. However, this metaphor for the PSC algorithm does not converge naturally, the algorithm must use an artificial parameter, called inertia term (α), to ensure convergence. This paper proposes a simple modification to the PSC algorithm, resulting in the Modified Particle Swarm Clustering (mPSC) algorithm, by modifying the metaphor of human social order to eliminate the artificial parameter of the system and, consequently, the memory of the particles' velocity. © 2010 IEEE.
Descrição
Palavras-chave
Assuntos Scopus
Artificial parameters , Flock of Birds , Memoryless , Particle swarm , PSO(particle swarm optimization) , Search spaces , Simple modifications , Social animals , Social order