Abstract
A good neural network has to produce precise results not only when dealing with training data but also when it is used for cases which has not been presented during a process of training. It means that in order to test a neural network’s performance you have to possess more examples than you use to train it. But what should you do if you have even a problem with gathering enough data for training? Where can you find new examples? This problem, in one of two possible kinds of overfitting, can be solved by using a method which will be presented in this article.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Arnold R. A. (1996) Economics. West Publishing Company, USA
Masters T. (1996) Practical Neural Network Recipes in C++. Scientific-Technical Publishing Company, Warsaw
Piegat A. (2001) Fuzzy Modeling and Control. A Springer — Verlag Company, Heidelberg - New York
Zielinski J. S. (2000) Intelligence systems in management, Theory and practice. Scientific Publishing Company PWN, Warsaw
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rejer, I., Piegat, A. (2003). How to Test Overfitting When There is not Enough Data for a Testing Set?. In: Rutkowski, L., Kacprzyk, J. (eds) Neural Networks and Soft Computing. Advances in Soft Computing, vol 19. Physica, Heidelberg. https://doi.org/10.1007/978-3-7908-1902-1_37
Download citation
DOI: https://doi.org/10.1007/978-3-7908-1902-1_37
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-7908-0005-0
Online ISBN: 978-3-7908-1902-1
eBook Packages: Springer Book Archive