Meta-conformity approach to reliable classification

Article type: Research Article

Authors: Smirnov, E.N.^a | Nalbantov, G.I.^a | Kaptein, A.M.^b

Affiliations: [a] Department of Knowledge Engineering, Faculty of Humanities and Sciences, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands. E-mail: smirnov@micc.unimaas.nl, g.nalbantov@micc.unimaas.nl | [b] Archives and Information Studies, University of Amsterdam, 1012 XT Amsterdam, The Netherlands. E-mail: a.m.kaptein@uva.nl

Abstract: The conformity framework has recently been proposed for the task of reliable classification. Given a classifier B, the framework allows to obtain p-values of the classifications assigned to individual instances. However, applying the framework is a difficult problem: we need to construct an instance non-conformity function for the classifier B. To avoid constructing such a function we propose a meta-conformity approach.11This paper is an extended version of [20]. If a conformity-based classifier M is available, the approach is to train M as a meta classifier that predicts the correctness of each classification of the classifier B. In this way the classification p-values of the classifier B are represented by the classification p-values of the classifier M. The meta-conformity approach can be used for constructing classifiers with predefined generalization performance. Experiments show that the approach results in classifiers that can outperform existing conformity-based classifiers.

Keywords: Reliable classification, conformity framework, meta classification

DOI: 10.3233/IDA-2009-0400

Journal: Intelligent Data Analysis, vol. 13, no. 6, pp. 901-915, 2009