Data Clustering Using Mother Tree Optimization

Wael Korani, Malek Mouhoub

Abstract: Clustering is the process of dividing data objects into different groups called clusters, without prior knowledge. Traditional clustering techniques might suffer from stagnation, where the solution is stuck in a local optimum. In the last decade, many metaheuristics, including swarm intelligence, have been applied to address the problem of clustering stagnation in a reasonable time. We propose a new clustering framework that is based on metaheuristics and, more precisely, swarm intelligence optimization algorithms that include particle swarm optimization (PSO) (Kennedy and Eberhart, 1995), whale optimization algorithm (WOA) (Mirjalili and Lewis, 2016), bacterial foraging optimization algorithm (BFOA) (Das et al., 2009) and mother tree optimization (MTO). To evaluate the performance of our framework and the new metaheuristic based on MTO called CMTO, we conducted a set of experiments on eight different datasets and using four different metrics: rand coefficient, Jaccard coefficient, d