Using weighted dynamic classifier selection methods in ensembles with different levels of diversity

Issue title: Hybrid Intelligent systems in Ensembles

Article type: Research Article

Authors: Canuto, Anne M.P.^{; *} | Fagundes, Diogo | Abreu, Marjory C.C. | Junior, João C. Xavier

Affiliations: Informatics and Applied Mathematics Department, Federal University of Rio Grande do Norte UFRN, Natal, RN, Brazil, 59072-970

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

Abstract: There are two main approaches to combine the output of classifiers within a multi-classifier system (MCS), which are: combination-based and selection-based methods. In selection-based methods, only one classifier is needed to correctly classify the input pattern. The choice of a classifier is typically based on the certainty of the current decision. The use of weights can be very useful for the final decision of a selection-based MCS since it can provide a confidence degree for each classifier. This paper presents the use of two confidence measures applied in three selection-based methods. The main aim of this paper is to analyze the benefits of using weights in the main selection-based methods and which confidence measure is more suitable to be used.

Keywords: Classifier ensembles, selection-based combination methods, confidence measures

DOI: 10.3233/HIS-2006-3303

Journal: International Journal of Hybrid Intelligent Systems, vol. 3, no. 3, pp. 147-158, 2006