This paper analyzes the application of a particular class of Bregman divergences to design cost-sensitive classifiers for multiclass problems. We show that these divergence measures can be used to estimate posterior probabilities with maximal accuracy for the probability values that are close to the decision boundaries. Asymptotically, the proposed divergence measures provide classifiers minimizing the sum of decision costs in non-separable problems, and maximizing a margin in separable MAP problems.
Bibliographical noteFunding Information:
Acknowledgements This work was partially funded by project TEC2008-01348 from the Spanish Ministry of Science and Innovation.
Copyright 2009 Elsevier B.V., All rights reserved.
- Bregman divergence
- Cost sensitive learning
- Maximum margin
- Posterior class probabilities