Multi-objective clustering ensemble

Issue title: Hybridization of Intelligent Systems

Guest editors: M. Köppenx and R. Webery

Article type: Research Article

Authors: Faceli, Katti^{a; *} | de Carvalho, André C. P. L. F.^a | de Souto, Marcílio C.P.^b

Affiliations: [a] ICMC – Universidade de São Paulo, 13560-970, São Carlos, SP, Brazil | [b] DIMAp – Universidade Federal do Rio Grande do Norte, 59072-970, Natal, RN, Brazil | [x] Kyushu Institute of Technology | [y] University of Chile

Correspondence: [*] Corresponding author. E-mail: [email protected]

Abstract: In this paper we present an algorithm for cluster analysis that integrates aspects from cluster ensemble and multi-objective clustering. The algorithm is constituted by a Pareto-based multi-objective genetic algorithm that uses clustering validation measures as the objective functions. This algorithm also uses a special consensus crossover operator. The proposed algorithm can deal with datasets with different types of clusters, without the need of much expertise in cluster analysis and the application domain. Moreover, it results in a concise and stable set of partitions representing different trade-offs between two validation measures related to different clustering criteria.

DOI: 10.3233/HIS-2007-4302

Journal: International Journal of Hybrid Intelligent Systems, vol. 4, no. 3, pp. 145-156, 2007