Check Swinburne's subscriptions for full text availabilitySearch Swinburne Research Bank
Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.3/26095
- Title
- A cascade of neural networks for complex classification
- Author(s)
- Philpot, David; Hendtlass, Tim
- Abstract
- Cascades of neural networks (CONN) is a technique that uses several different neural networks to find a relationship in a set of training data. It can be thought of as a divide and conquer technique. Each neural network learns a particular region in the training space, and all the learned regions of all the networks fit together to cover the entire training space in a similar way to pieces of a jigsaw puzzle.
- Publication type
- Conference paper
- Research centre
- Swinburne University of Technology. School of Biophysical Sciences and Electrical Engineering. Centre for Intelligent Systems
- Source
- Proceedings 9th International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems (IEA/AIE 96), Fukuoka, Japan, 04-07 June 1996, p. 791
- Publication year
- 1996
- Keyword(s)
- Cascades of neural networks; CONN; Neural networks
- Publisher
- Gordon and Breach Science Publishers
- ISBN
- 9056995243
- Publisher URL
- http://www.peppersoftware.com/dphilpot/publications.html
- Copyright
- Copyright © 1996 by OPA (Overseas Publishers Association) Amsterdam B. V. Published in The Netherlands under license by Gordon and Breach Science Publishers SA. All rights reserved.
- Peer reviewed
