Sampling scheme for better RBF network
Hyontai Sug · 2009
Neural networks have been developed for machine learning and data mining tasks, and because data mining problems contain a large amount of data, sampling is a necessity for the success of the task. For this reason, this paper suggests an effective sampling technique that is based on a generated decision tree, where the trees are generated based on a fast and dirty tree generation algorithm. Experiments with several sample sizes and RBF network showed that the method is more effective with respect to accuracy than conventional random sampling method.