Browse by author
Lookup NU author(s): Tom Musicka, Professor Elaine Martin
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Bayesian methods offer a sound mathematical foundation on which to optimise models, prevent overfitting of data and realise the incorporation of prior knowledge when building models. The standard Adaptive Network-based Fuzzy Inference System (ANFIS) is modified to enable the integration of Bayesian methodologies. The proposed approach is compared with linear partial least squares, neural networks and a standard gradient descent based ANFIS model for the prediction of a mechanical property in steel manufacturing.
Author(s): Musicka T, Martin E, Morris J, Kitson P
Editor(s): McNulty, G.J.
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: Quality, Reliability and Maintenance (QRM)
Year of Conference: 2002
Pages: 195-198
Publisher: Wiley
Library holdings: Search Newcastle University Library for this item
ISBN: 9781860583698
