ABSTRACT
In this chapter we describe neuro-fuzzy systems which combine the advantages of numerical computations of neural networks with symbolic processing of fuzzy sets. First, we give a brief introduction to fuzzy sets, sufficient to understand the topics covered in the chapter. This includes a discussion of methods for eliciting membership functions. Next, several typical neuro-fuzzy algorithms are discussed and illustrated. The last few sections concentrate on fuzzy neural networks, where basic processing components (fuzzy neurons) and several general architectures are discussed. In particular, it is shown that some topologies of the networks, such as logic processors, can be exploited in a logic-based approximation of functional relationships.
