A Partitional Cooperative Coevolutionary Group Search Optimization Approach for Data Clustering
Luciano D. S. Pacífico, Teresa B. Ludermir · 2019
Clustering algorithms are important methods for exploratory data analysis with application in many fields, such as data mining, image understanding, text analysis, and so on. Partitional clustering models are the most popular clustering methods, but these approaches suffer from some limitations, like the sensibility to algorithm initialization and the lack of mechanisms to help them escaping from local minima points. In this work, a new partitional cooperative coevolutionary algorithm based on Group Search Optimization (GSO) is proposed (CCGSO) to deal with clustering task. Evolutionary Algorithms (EAs), such as GSO, are nature-inspired meta-heuristics known for their capabilities to find global optima solutions even when dealing with hard optimization problems. However, many EAs are based on competitive behavior among population individuals, although it is known that sometimes cooperation may result in better solutions then sheer competition. Experiments are executed on real benchmark data sets to evaluate the performance of proposed algorithm in comparison to other well-known competitive partitional evolutionary clustering methods from literature. Experimental results show that CCGSO is able to achieve better average performances when dealing with clustering task than competitive evolutionary clustering approaches.