The Performance Analysis of a Novel Enhanced Artificial Bee Colony Inspired Global Best Harmony Search Algorithm for Clustering

Abstract

Clustering is the unsupervised classification of data items of patterns into groups, each of which should be as homogeneous as possible. The problem of clustering has been addressed in many contexts in many disciplines and this reflects its broad appeal and usefulness in exploratory data analysis. This paper presents a new clustering algorithm, called GHSBEEK which is a combination of the Global best Harmony search (GHS) with features of Artificial Bee Colony (ABC) and K-means algorithms. Global-best Harmony search (GHS) is a derivative-free optimization algorithm, which draws inspiration from the musical process of searching for a perfect state of harmony. It has a remarkable advantage of algorithm simplicity. However, it suffers from a slow search speed. The ABC algorithm is applied to improve the members of the Harmony Memory based on their fitness values and hence improves the convergence rate of the Harmony Search method. The GHSBEEK algorithm has been used for data clustering on several benchmark data sets. The clustering performance of the proposed algorithm is compared with the GHS, PSO, and K-means. The simulation results show that the proposed algorithm outperforms the other algorithms in terms of accuracy, robustness, and convergence speed.