论文信息 - Fuzzy RuleNet: an artificial neural network model for fuzzy classification

Fuzzy RuleNet: an artificial neural network model for fuzzy classification

Fuzzy RuleNet is a new Artificial Neural Network model for fuzzy classification. Fuzzy RuleNet allows the translation of symbolic knowledge into the network and viceversa. Therefore an initialization of th~ network with a given set of fuzzy rules and the corresponding membership functions is possible. It is also possible to represent the knowledge contained in the network as a fuzzy rule based system. Fuzzy RuleNet is a feedforward network model with a supervised learning algorithm and a dynamic architecture. As the number of the layers is fixed and the number of nodes in each layer is determined by the learning algorithm, the network architecture is exactly defined by the problem. RuleNet [11] and Fuzzy RuleNet are extensions to M-RCE networks presented in [I0].

Nadine N. Tschichold-Gürman

[1] Peter E. Hart,et al. Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.

[2] H. Zimmermann,et al. Fuzzy Set Theory and Its Applications , 1993 .

[3] Casimir A. Kulikowski,et al. Computer Systems That Learn: Classification and Prediction Methods from Statistics, Neural Nets, Machine Learning and Expert Systems , 1990 .

[4] Hans-Jürgen Zimmermann,et al. Fuzzy Set Theory - and Its Applications , 1985 .

[5] Stuart L. Crawford. Extensions to the CART Algorithm , 1989, Int. J. Man Mach. Stud..

[6] Patrick K. Simpson. Fuzzy min-max classification with neural networks , 1991, [1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering.

[7] Christopher L. Scofield,et al. Neural networks and speech processing , 1991, The Kluwer international series in engineering and computer science.

[8] Steven L. Salzberg,et al. Learning with Nested Generalized Exemplars , 1990 .