Please use this identifier to cite or link to this item: http://hdl.handle.net/2381/42416
Title: Generation of explicit knowledge from empirical data through pruning of trainable neural networks
Authors: Gorban, Alexander N.
Mirkes, Eugeniy M.
Tsaregorodtsev, Victor G.
First Published: 6-Aug-2002
Presented at: International Joint Conference on Neural Networks IJCNN '99, Washington, DC, USA
Start Date: 10-Jul-1999
End Date: 16-Jul-1999
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Citation: Proceedings of the International Joint Conference on Neural Networks IJCNN '99, 1999, 6, pp. 4393-4398
Abstract: This paper presents a generalized technology of extraction of explicit knowledge from data. The main ideas are 1) maximal reduction of network complexity (not only removal of neurons or synapses, but removal all the unnecessary elements and signals and reduction of the complexity of elements), 2) using of adjustable and flexible pruning process (the pruning sequence shouldn't be predetermined - the user should have a possibility to prune network on his own way in order to achieve a desired network structure for the purpose of extraction of rules of desired type and form), and 3) extraction of rules not in predetermined but any desired form. Some considerations and notes about network architecture and training process and applicability of currently developed pruning techniques and rule extraction algorithms are discussed. This technology, being developed by us for more than 10 years, allowed us to create dozens of knowledge-based expert systems.
DOI Link: 10.1109/IJCNN.1999.830876
ISSN: 1098-7576
ISBN: 0-7803-5529-6
Links: https://ieeexplore.ieee.org/document/830876/
http://hdl.handle.net/2381/42416
Version: Post-print
Status: Peer-reviewed
Type: Journal Article
Rights: Copyright © 2002, IEEE. Deposited with reference to the publisher’s open access archiving policy. (http://www.rioxx.net/licenses/all-rights-reserved)
Appears in Collections:Conference Papers & Presentations, Dept. of Mathematics

Files in This Item:
File Description SizeFormat 
0307083v1.pdfPost-review (final submitted author manuscript)131.66 kBAdobe PDFView/Open


Items in LRA are protected by copyright, with all rights reserved, unless otherwise indicated.