A Unified Framework of Binary Classifiers Ensemble for Multi-class Classification
We present a novel methods for multi-class classification by ensemble of binary classifiers for multi-class classification. The proposed method is characterized by a minimization problem of weighted divergences, and includes a lot of conventional methods as special cases. We discuss relationship between the proposed method and conventional methods and statistical properties of the proposed method. A small experiment shows that the proposed method can effectively incorporate information of multiple binary classifiers into multi-class classifier.
KeywordsEnsemble learning Mixture of divergences
Unable to display preview. Download preview PDF.
- 2.Amari, S., Nagaoka, H.: Methods of Information Geometry. Translations of Mathematical Monographs, vol. 191. Oxford University Press (2000)Google Scholar
- 9.Takenouchi, T., Ishii, S.: Ternary bradley-terry model-based decoding for multi-class classification and its extensions. Machine Learning, 1–24 (2011)Google Scholar
- 10.Weston, J., Watkins, C.: Support vector machines for multi-class pattern recognition. In: Proceedings of the Seventh European Symposium on Artificial Neural Networks, vol. 4(6) (1999)Google Scholar
- 11.Zadrozny, B.: Reducing multiclass to binary by coupling probability estimates. In: Dietterich, T.G., Becker, S., Ghahramani, Z. (eds.) Advances in Neural Information Processing Systems, vol. 14. MIT Press, Cambridge (2001)Google Scholar