Particles with Age for Data Clustering

This paper proposes a novel particle swarm optimisation (PSO) algorithm using the concept of age of particles. Effective fitness of a particle depends both on its functional value and age. Age of a newly generated particle is taken as zero, and in every iteration age of each individual is increased by one. In this paper, a trapezoidal aging function is considered. The model aims to emulate natural swarm system in a more natural way. The effectiveness of this concept is demonstrated by cluster analysis. Results show that the model provides enhanced performance and maintains more diversity in the swarm and thereby allows the particles to be robust to trace the changing environment.