A Framework for Estimating Performance Improvements in Hybrid Pattern Classifiers

Abstract

Classiication methods often perform signiicantly below Bayesian limits in complex , high-dimensional classiication tasks because of model bias, inadequate training data and noise/variability in the data. When several classiiers are used for a given task, selecting one method over all others discards potentially valuable information. Strategies aimed at… (More)

Topics

Cite this paper

@inproceedings{Tumer1994AFF, title={A Framework for Estimating Performance Improvements in Hybrid Pattern Classifiers}, author={Kagan Tumer and Joydeep Ghosh}, year={1994} }