Patent
1991-09-09
1995-08-01
Knepper, David D.
395 21, 395 24, 395 27, 395 50, G06F 1518
Patent
active
054386446
ABSTRACT:
A rule-based expert system is generated from a neural network. The neural network is trained in such a way as to avoid redundancy and to select input weights to the various processing elements in such a way as to nullify the input weights which have smaller absolute values. The neural network is translated into a set of rules by a heuristic search technique. Additionally, the translation distinguishes between positive and negative attributes for efficiency and can adequately explore rule size exponential with a given parameter. Both explicit and implicit knowledge of adapted neural networks are decoded and represented as if--then rules.
REFERENCES:
patent: 4730259 (1988-03-01), Gallant
patent: 4803642 (1988-02-01), Muranaga
patent: 4860213 (1989-08-01), Bonissone
patent: 4958939 (1990-09-01), Samad
patent: 5014219 (1991-05-01), White
patent: 5052043 (1991-09-01), Gaborski
patent: 5092343 (1992-03-01), Spitzer et al.
patent: 5113485 (1992-05-01), Horan et al.
patent: 5136686 (1992-08-01), Koza
patent: 5214745 (1993-05-01), Sutherland
patent: 5218529 (1993-06-01), Meyer et al.
patent: 5226092 (1993-07-01), Chen
patent: 5241620 (1993-08-01), Ruggiero
patent: 5251268 (1993-10-01), Colley et al.
Karen Okagaki; "Using Expert Systems to Validate and Verify Neural Networks"; Inter. Neural Network Conf. (INNC) Jul. 9-13, 1990; pp. 314-317.
"Blackboard Systems-Blackboard Application Systems, Blackboard System from a Knowledge Engineering Perspective"-H. Penny Nii, The Ai Magazine, pp. 82-106, Aug. 1986.
"An Empirical Comparison of ID3 and Back-propagation"-Douglas Fisher and Kathleen B. McKusick, Machine Learning, pp. 788-793.
"An Experimental Comparison of Symbolic and Connectionist Learning Algorithms", Raymond Mooney, Jude Shavlik, Geoffrey Towell, Alan Gove, Machine Learning, pp. 775-780.
"Connectionist Expert Systems", Stephen I. Gallant, Communications of the ACM, vol. 31, No. 2, pp. 152-168, Feb. 1988.
"Learning Internal Representations by Error Propagation" D. E. Rumelhart, G. E. Hinton, and R. J. Williams, Learning Internal Representations, pp. 319-362.
"Enhancing Performance of Expert Systems by Automated Discovery by Meta-Rules", Li-Min Fu and Bruce G. Buchanan, pp. 107-115, 1984, The First Conference on Artificial Intelligence Application.
"Learning Intermediate Concepts in Constructing a Hierarchical Knowledge Base"-Li-Min Fu and Bruce G. Buchanan, pp. 659-666, The 9th International Joint Conference on Artificial Intelligence, 1985.
"Adaptive Signal Detection in Noisy Environments", Li-Min Fu, Journal of Neural Network Computing, pp. 42-50, Spring 1990.
"Simplifying Decision Trees", J. R. Quinlan, Man-Machine Studies, pp. 221-234, 1987.
"On Debugging Rule Sets When Reasoning Under Uncertainty"-David C. Wilkins and Bruce G. Buchanan, Science/Learning, pp. 448-454.
Hafiz Tariq R.
Knepper David D.
University of Florida
LandOfFree
Translation of a neural network into a rule-based expert system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Translation of a neural network into a rule-based expert system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Translation of a neural network into a rule-based expert system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-739113