PSO driven collaborative clustering: a clustering algorithm for ubiquitous environments

Abstract

The goal of this article is to introduce two existing clustering approaches into the domain of ubiquitous knowledge discovery. First we demonstrate how horizontal collaborative clustering can be performed in a ubiquitous environment and discuss the ability of these two clustering techniques to cope with privacy constraints. Next, we illustrate how a particle swarm optimization driven version of this clustering algorithm can be used in KDUbiq research and we introduce a ¯tness functions whose objective is to ¯nd similar cluster composition across data locations. Finally, we run an experiment which shows the potential of PSO driven collaborative clustering in a ubiquitous environment with privacy issues