Dynamic ensemble selection for multi-class classification with one-class classifiers. (November 2018)

• Record Type: Journal Article
• Authors: Krawczyk, Bartosz; Galar, Mikel; Woźniak, Michał; Bustince, Humberto; Herrera, Francisco
• Is Part Of: Pattern recognition. Volume 83(2018:Nov.)
• Page Start: 34
• Page End: 51
• Publication Date: 2018-11
• Subjects: Machine learning -- Classifier ensemble -- One-class classification -- Multi-class decomposition -- Dynamic classifier selection -- Ensemble pruning; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.05.015
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Dynamic ensemble selection for multi-class decomposition with one-class classifiers. Efficient framework for difficult data with high number of classes. Removal of non-competent classifiers from decision making process. Threshold-based selection mechanism for further ensemble pruning Extensive experimental study backed-up with statistical analysis. Abstract: In this paper we deal with the problem of addressing multi-class problems with decomposition strategies. Based on the divide-and-conquer principle, a multi-class problem is divided into a number of easier to solve sub-problems. In order to do so, binary decomposition is considered to be the most popular approach. However, when using this strategy we may deal with the problem of non-competent classifiers. Otherwise, recent studies highlighted the potential usefulness of one-class classifiers for this task. Despite not using all the available knowledge, one-class classifiers have several desirable properties that may benefit the decomposition task. From this perspective, we propose a novel approach for combining one-class classifiers to solve multi class problems based on dynamic ensemble selection, which allows us to discard non-competent classifiers to improve the robustness of the combination phase. We consider the neighborhood of each instance to decide whether a classifier may be competent or not. We further augment this with a threshold option that prevents from the selection of classifiers corresponding to classes with too little examples in this neighborhood. To evaluate the usefulness of our approach an extensive experimental study is carried out, backed-up by a thorough statistical analysis. The results obtained show the high quality of our proposal and that the dynamic selection of one-class classifiers is a useful tool for decomposing multi-class problems.