Skip to Main Content
A hybrid approach for designing fuzzy rule-based systems based on clustering and a class of fuzzy neural networks is introduced. Firstly, an unsupervised clustering technique is used to determine the number of fuzzy rules and generate an initial fuzzy rule base from the given input-output data. Secondly, a class of fuzzy neural networks is constructed and its weights are tuned to make the parameters of the constructed fuzzy rule base more precise. Finally, we focus on function approximation problems as a vehicle to evaluate its performance.