The paper examines the general classifier combination problem under strict separation of the classifier and combinator design. Several desirable combinator properties are identified: omnitype mixed type and correlated classifier combination, redundant classifier elimination, model complexity control, and dynamic selection combination. By adapting some of the theories and algorithms developed for neural network learning. They present a combination model which provides a solution to these problems. Experimental results on handwritten digits verify these findings.
A theory of classifier combination: the neural network approach
1995-01-01
525622 byte
Conference paper
Electronic Resource
English
A Theory of Classifier Combination: The Neural Network Approach
British Library Conference Proceedings | 1995
|Dynamic classifier combination using neural network [2422-04]
British Library Conference Proceedings | 1995
|Spiking Neural Network E-Nose classifier chip
IEEE | 2010
|COMPOSITE MATERIALS' CONDITION CLASSIFIER BASED ON NEURAL NETWORK OF ADAPTIVE RESONANCE THEORY
DOAJ | 2012
|Classifier Combination for Vehicle Silhouettes Recognition
British Library Conference Proceedings | 1999
|