Estimation of Performance Bounds in Supervised Classification
Pierre Comon, Jean-Luc Voz, Michel Verleysen · 1994
The Bayes theory gives the ultimate performances that can be reached in a classification problem. We present in this paper a method that allows to estimate these performance bounds given any finite data set, by building a classifier based on two successive estimations of probability densities, which asymptotically converge to the optimal Bayesian classifier.