Skip to Main Content
This paper proposes a heuristic method to extract candidates of support vector from training set. Training a support vector on the extracted candidates, we attain good generalization on test set. It shows that candidates of support vector contain almost all the necessary information to solve a given classification task. This method is also applied to incorporate prior knowledge into support vector machines. Experiments on digits recognition show the same performance as virtual support vector method.
Machine Learning and Cybernetics, 2003 International Conference on (Volume:5 )
Date of Conference: 2-5 Nov. 2003