Naive Bayesian classifiers with extreme probability features

L.C. van der Gaag, A. Capotorti

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Despite their popularity, naive Bayesian classifiers are not well suited for real-world applications involving extreme probability features. As will be demonstrated in this paper, methods used to forestall the inclusion of zero probability parameters in naive classifiers have quite counterintuitive effects. An elegant, principled solution for handling extreme probability events is available however, from coherent conditional probability theory. We will show how this theory can be integrated in standard naive Bayesian classifiers, and then present a computational framework that retains the classifiers’ efficiency in the presence of a limited number of extreme probability features.
Original languageEnglish
Title of host publication Proceedings of Machine Learning Research
Pages499-510
Volume72
Publication statusPublished - 2018

Publication series

NameProceedings of Machine Learning Research
Volume72
ISSN (Electronic)1938-7228

Keywords

  • Naive Bayesian classifiers
  • Extreme probabilities
  • Coherent conditional probabilitytheory
  • Computational efficiency

Fingerprint

Dive into the research topics of 'Naive Bayesian classifiers with extreme probability features'. Together they form a unique fingerprint.

Cite this