Hostname: page-component-848d4c4894-5nwft Total loading time: 0 Render date: 2024-05-16T22:37:30.981Z Has data issue: false hasContentIssue false
  • English
  • Français

The use of neural networks to derive prediction relationships for feed composition from near-infra-red monochromator scans

Published online by Cambridge University Press:  21 November 2017

E A Chadoulos ,
J M Bishop ,
A T Chamberlain  and
C Piotrowski
E A Chadoulos
Affiliation:
Department of Cybernetics
J M Bishop
Affiliation:
Department of Cybernetics
A T Chamberlain
Affiliation:
Department of Agriculture, University of Reading, Reading
C Piotrowski
Affiliation:
NIR Services, J Bibby Agriculture Ltd, Oxford Road, Adderbury, Banbury
Get access

Extract

The derivation of predictive relationships from near-infrared scan results is generally performed using regression analysis but for this technique relationships must be a continuous function. Neural networks allow non-functional relationships to be investigated and in other fields have improved prediction quality by 10% - 25%.

Neural networks are an artificial intelligence tool and can be represented as a multi-layered network of nodes connected by vectors (Figure 1), each node receiving one or more inputs which are transformed and combined and then passed on to one or more nodes in the next layer.

Type
Ruminant Nutrition and Digestion
Information
Proceedings of the British Society of Animal Production (1972) , Volume 1994 , March 1994 , pp. 92
Copyright
Copyright © The British Society of Animal Production 1994

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

McClelland, J.L. and Rumelhart, D.E. 1989. Explorations in parallel distributed processing. MIT Press, USA.Google Scholar