An experimental decision of samples for RBF neural networks

Authors:
Hyontai Sug
Affiliations:
Division of Computer and Information Engineering, Dongseo University, Busan, Republic of Korea
Venue:
MUSP'09 Proceedings of the 9th WSEAS international conference on Multimedia systems & signal processing
Year:
2009

Citing 8
Cited 1

Effects of Sample Size in Classifier Design

IEEE Transactions on Pattern Analysis and Machine Intelligence
Small Sample Size Effects in Statistical Pattern Recognition: Recommendations for Practitioners

IEEE Transactions on Pattern Analysis and Machine Intelligence
Efficient progressive sampling

KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Discretization: An Enabling Technique

Data Mining and Knowledge Discovery
Introduction to Data Mining, (First Edition)

Introduction to Data Mining, (First Edition)
2008 Special Issue: Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance

Neural Networks
2008 Special Issue: Classifier performance estimation under the constraint of a finite sample size: Resampling schemes applied to neural network classifiers

Neural Networks

Designing neural networks for tackling hard classification problems

WSEAS TRANSACTIONS on SYSTEMS

Quantified Score

Hi-index	0.00

Visualization

Abstract

It is not known to decide a proper sample size for data mining tasks, so the task of deciding proper sample sizes for RBF neural networks that are one of the important data mining algorithms tend to be arbitrary. In RBF networks as the size of samples grows, the improvement in error rate becomes better slowly. But we cannot use larger and larger samples, because there are some fluctuations in accuracy as the sample size grows. This paper suggests an objective approach in determining proper samples to find good RBF networks with respect to accuracy. Experiments with two relatively large data sets showed very promising results.