ABSTRACT
Evolutionary algorithms have become more popular nowadays in order to solve non-linear complex real world problems. One of these many applications includes clustering of complex data into numerous small groups, which have certain similarities. The process of clustering may be achieved using certain measures such as minimum intra-cluster variance or maximum inter-cluster variance, depending upon the kind of data at hand. For such clustering problems, population-based evolutionary algorithms can be utilized to perform a clustering process using various performance measures as an objective function. These algorithms take advantage of a random population-based stochastic approach and treats clustering as a means of determining an optimal solution such as in document clustering, where the optimum cluster formation is achieved on the given data. Based on the nature of the data, these approaches have applications in various domains, such as text document clustering, image segmentation, customer segmentation, telecommunication clustering, healthcare analysis, pollution effect analysis etc. Among the popular examples of evolutionary approaches are Genetic Algorithm, Gene Expression Programming, Differential Evolution, Particle Swarm Optimization, Gray Wolf Optimizer, Antlion Optimizer, Neuroevolutionary Algorithm etc. The chapter presents the state-of-the-art stock of various evolutionary techniques for clustering, their applications in different domains, and key advantages and disadvantages.
