Data Driven Fuzzy Modelling with Neural Networks
Extraction of models for complex systems from numerical data of behavior is studied. In particular, systems representable as sets of fuzzy if-then rules where the premises are not connected by t-norms, but by a compensating aggregation operator are discussed. A method is presented to extract this kind of fuzzy rules with support of neural networks. Finally it is shown that it is possible to extract compensating fuzzy if-then rules from a great number of already existing feedforward neural networks.
KeywordsNeural Network Hide Layer Fuzzy Logic Fuzzy Rule Fuzzy Inference System
Unable to display preview. Download preview PDF.