Title :
Rule extraction using a novel class of fuzzy degraded hyperellipsoidal composite neural networks
Author :
Su, Mu-Chun ; Kao, Chien-Jen ; Liu, Kai-Ming ; Liu, Chi-Yeh
Author_Institution :
Dept. of Electr. Eng., Tamkang Univ., Tamsui, Taiwan
Abstract :
Presents an innovative approach to rule extraction directly from experimental numerical data for system identification. The authors discuss how to use a novel class of fuzzy degraded hyperellipsoidal composite neural networks (FDHECNN´s) to extract fuzzy if-then rules. The fuzzy rules are defined by hyperellipsoids of which principal axes are parallel to the coordinates of the input space. These rules are extracted from the parameters of the trained FDHECNN´s. Based on a special learning scheme, the FDHECNN´s can evolve automatically to acquire a set of fuzzy rules for approximating the input/output functions considered systems. A highly nonlinear system is used to test the proposed neuro-fuzzy systems
Keywords :
fuzzy logic; fuzzy neural nets; identification; learning (artificial intelligence); fuzzy degraded hyperellipsoidal composite neural networks; fuzzy if-then rules; input/output functions approximation; learning scheme; rule extraction; system identification; Backpropagation; Clustering algorithms; Data mining; Degradation; Fuzzy neural networks; Fuzzy systems; Humans; Input variables; Knowledge acquisition; Neural networks;
Conference_Titel :
Fuzzy Systems, 1995. International Joint Conference of the Fourth IEEE International Conference on Fuzzy Systems and The Second International Fuzzy Engineering Symposium., Proceedings of 1995 IEEE Int
Conference_Location :
Yokohama
Print_ISBN :
0-7803-2461-7
DOI :
10.1109/FUZZY.1995.409686
