A new method to evaluate a trained artificial neural network

Date

2001-01-01

Advisors

Journal Title

Journal ISSN

ISSN

Volume Title

Publisher

Type

Other

Peer reviewed

Abstract

Description

It is possible for a trained neural network to give a false mapping. We propose a new approach to evaluate a trained neural network. A new parameter is defined to identify the different potential roles of the individual input factors based on the trained connections of the nodes in the network. Compared with field-specific knowledge, the dominance of individual input factors can be checked and then false mappings satisfying only the specific data set may be avoided. In this way, the available data could be fully applied to the training stage and the validation is simple and efficient.

Keywords

RAE 2008, UoA 23 Computer Science and Informatics, neural nets

Citation

Yang, Y., Hinde, C.J., and Gillingwater, D. (2001) A new method to evaluate a trained artificial neural network. In: Proceedings. IJCNN '01. International Joint Conference Neural Networks, Washington, DC, 15-19 July, Vol.4, pp. 2620-2625

Rights

Research Institute

Institute of Artificial Intelligence (IAI)