Grey Wolf Optimizer (GWO) Algorithm to Solve the Partitional Clustering Problem

Abstract

The clustering which is an unsupervised classification method is very important for data processing applications. The main purpose of the clustering is to separate the data samples into different groups by using the similarity (or dissimilarity) between data samples. There are many conventional and heuristic algorithms which are used for the clustering problem. Nevertheless, in last years, it is seen that many new techniques are proposed and improved to solve the clustering problem. In this paper, grey wolf optimization (GWO) algorithm which is modelled according to the social behavior of grey wolves is applied to partition the data samples by searching the optimal center of the clusters. The clustering performance of the GWO is compared with the performances of the three clustering algorithms: k-means, kmedoids and fuzzy c-means algorithms. The experiments show that the GWO algorithm has generally better results than the other clustering algorithms and can be alternatively applied on the clustering problem.

An Approach Based on Tunicate Swarm Algorithm to Solve Partitional Clustering Problem

“…In the phase of clustering process, the data instances are divided into some sub groups according to their similar and dissimilar attributes. Eventually, the main goal of the data clustering is to obtain some homogeneous sets [3,4,45]. Partitional clustering methods are distributed the N data instances to k clusters (sets) in accordance with the similarity of the data instances.…”

“…The fitness value of K-Means is calculated with Equation (1). The value of the each center ( ) is calculated with Equation (2) [4].…”

“…When a cluster center is determined in K-Means algorithm, any location in upper and lower bound can be a center of cluster. In K-Medoids method the best medoids are selected as cluster centers and a medoid should be an instance from all dataset [4,51]. In this study, the Partitioning Around Medoids (PAM) [24] is selected as K-Medoids algorithm.…”

“…The intention of the clustering algorithms is to split instances into some groups according to their similar and dissimilar features. If one of the clustering technique is implemented on any dataset, the dataset separated to some groups and it is expected that the similarities of these groups will be in a minimum level [3,4]. The main goal of the using the previous data samples is to obtain a conclusion, estimate a future statement and in diagnosis process [2].…”

“…In recently, optimization-based metaheuristic algorithms are used for many different optimization problems, because of problem free, has a simple structure and easy adaptable to any optimization problems [36]. Some of these metaheuristic algorithms for solving data clustering problems are such as Genetic Algorithm (GA) [37,38], Teacher Learning Based Optimization (TLBO) [39], Ant Colony Optimization (ACO) [40], Artificial Bee Colony (ABC) [3,41], Gravitational Search Algorithm (GSA) [42], Particle Swarm Optimization (PSO) [43], Grey Wolf Optimizer (GWO) [4] and Cuckoo Search (CS) [44,45] algorithms.…”