An Uncertainty Framework for Classification
Loo-Nin Teow, Kia-Fock Loe
We define a generalized likelihood function based on uncertainty measures and show that maximizing such a likelihood function for different measures induces different types of classifiers. In the probabilistic framework, we obtain classifiers that optimize the cross-entropy function. In the possibilistic framework, we obtain classifiers that maximize the interclass margin. Furthermore, we show that the support vector machine is a sub-class of these maximum-margin classifiers.
Keywords: uncertainty framework, classification, likelihood, probabilistic, possibilistic, maxi
PS Link: http://www.comp.nus.edu.sg/~teowloon/uai2000_teow.ps
PDF Link: /papers/00/p574-teow.pdf
AUTHOR = "Loo-Nin Teow
and Kia-Fock Loe",
TITLE = "An Uncertainty Framework for Classification",
BOOKTITLE = "Proceedings of the Sixteenth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-00)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "2000",
PAGES = "574--579"