Klawonn, Frank and Angelov, Plamen (2006) Evolving extended naive Bayes classifiers. In: Data Mining Workshops, 2006. ICDM Workshops 2006. Sixth IEEE International Conference on :. IEEE, Hong Kong, pp. 643-647. ISBN 0-7695-2702-7
Abstract
Naive Bayes classifiers are a very simple, but often effective tool for classification problems, although they are based on independence assumptions that do not hold in most cases. Extended naive Bayes classifiers also rely on independence assumptions, but break them down to artificial subclasses, in this way becoming more powerful than ordinary naive Bayes classifiers. Since the involved computations for Bayes classifiers are basically generalised mean value calculations, they easily render themselves to incremental and online learning. However, for the extended naive Bayes classifiers it is necessary, to choose and construct the subclasses, a problem whose answer is not obvious, especially in the case of online learning. In this paper we propose an evolving extended naive Bayes classifier that can learn and evolve in an online manner (c) IEEE Press