Title: An empirical study on multi-objective genetic algorithms using clustering techniques
Authors: M. Anusha; J.G.R. Sathiaseelan
Addresses: Department of CS, Bishop Heber College, Trichy, TN, India ' Department of CS, Bishop Heber College, Trichy, TN, India
Abstract: Clustering is a data mining technique widely used to find similar group of data. A better cluster always have most similar data while the elements from the different clusters are dissimilar. Genetic algorithms (GAs) are considered as a global searching technique for optimisation problems. In the recent years there are many conflicting measure of objectives which are need to be optimised concurrently to achieve a tradeoff. Traditionally, evolutionary algorithms (EAs) were used to solve single objective problems. Optimum performance in single objective optimisation often results low, when the situation deals with more than one objective. This situation creates a bottleneck for an alternate technique called as multi-objective optimisation using genetic algorithms which aids to find more solutions in data mining domain.
Keywords: cluster ensemble; independent objective-based approach; elitist selection strategy; multi-objective optimisation; genetic algorithms; clustering; data mining.
International Journal of Advanced Intelligence Paradigms, 2016 Vol.8 No.3, pp.343 - 354
Received: 10 Nov 2014
Accepted: 04 May 2015
Published online: 04 Jul 2016 *