Abstract
Methodology is the study of a whole body of methods used in a particular branch of activity. By this definition, we will not attempt to cover the complete methodology of applying neural networks to industrial applications, as we are not going to concern ourselves with the more detailed aspects of the implementation of neural networks. Thus we will not attempt to give more than the barest of statements concerning the detailed structure of nets: the numbers of nodes to use, the best learning algorithms, the most appropriate activation functions etc. We believe this would be unwise for a number of reasons: the detailed discussions derived from 3 years research and many different applications would be far too long; and, more significantly, we feel that any such conclusions would date rather rapidly as new techniques arise in the literature.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1992 ECSC — EEC — EAEC, Brussels — Luxembourg
About this chapter
Cite this chapter
Croall, I.F., Mason, J.P. (1992). Methodology. In: Croall, I.F., Mason, J.P. (eds) Industrial Applications of Neural Networks. Research Reports ESPRIT, vol 1. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-84837-7_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-84837-7_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-55875-0
Online ISBN: 978-3-642-84837-7
eBook Packages: Springer Book Archive