Title :
Improved rule generation for a neuro-fuzzy network
Author :
van Vuuren, P.A. ; Hoffman, A.J.
Author_Institution :
Potchefstroom Univ. for CHE, South Africa
Abstract :
The success of a neuro-fuzzy network is influenced by both its architecture and its learning algorithm. Currently, C.-J. Lin and C.-T. Lin´s (1997) FALCON-ART algorithm ranks amongst the best structure/parameter learning algorithms yet devised. The FALCON-ART algorithm is adapted for use in neuro-fuzzy networks responsible for pattern recognition tasks. In contrast with FALCON-ART, each cluster is issued with its own vigilance parameter. Consequently, the sizes of individual rule antecedents can be controlled. A fuzzy logic controller is employed for this purpose. When it was applied to the Iris recognition problem, the neuro-fuzzy network attained an average recognition rate of 95.07%. However, it fared slightly worse than a conventional neural network on a seismic discrimination task. The main advantages of the improved rule extraction algorithm are its speed, and the compactness of its resulting rule databases
Keywords :
ART neural nets; fuzzy control; fuzzy neural nets; knowledge based systems; learning (artificial intelligence); pattern recognition; FALCON-ART algorithm; Iris recognition problem; average recognition rate; fuzzy logic controller; individual rule antecedents; learning algorithm; neuro-fuzzy network; pattern recognition tasks; rule databases; rule extraction algorithm; rule generation; seismic discrimination task; structure/parameter learning algorithm; vigilance parameter; Automatic control; Clustering algorithms; Design optimization; Fuzzy control; Fuzzy neural networks; Neural networks; Parameter estimation; Pattern recognition; Size control; Subspace constraints;
Conference_Titel :
Systems, Man, and Cybernetics, 2000 IEEE International Conference on
Conference_Location :
Nashville, TN
Print_ISBN :
0-7803-6583-6
DOI :
10.1109/ICSMC.2000.884429
