Neural Network based Construction of Fuzzy Graphs
Michael R. Berthold, Klaus–Peter Huber · 1995
Function approximation using example data has gained considerable interest in the past. One interesting application is the approximation of the behaviour of simulation models, called metamodelling. The goal is to approximate the behaviour as well as to extract some understandable knowledge about the simulation model. In this paper a combination of a special type of Neural Network (Rectangular Basis Function Network) with a (de--)fuzzification module is used. The resulting system approximates real valued functions with an adjustable precision. A constructive algorithm builds the network from scratch, resulting in a structure where each hidden unit represents a rectangular area with a corresponding membership function (or a fuzzy point). The underlying knowledge can be extracted from the network in form of a Fuzzy Graph. I. Introduction The dynamic behaviour of systems can be analyzed using simulation models. Unfortunately, often the results are only available in the form of large datas...