Fuzzy approach to a neural network

A mathematical model of a multilevel, multilayer neural network has been developed which uses fuzzy theory for classification. The developed architecture allows imprecise or incomplete knowledge gained at one level of the network to be forwarded as input to the next level. The partial knowledge gained, which can be viewed as a set of fuzzy decisions, is mathematically represented in the network connection weights. The network first looks at the strongest features of the object and tries to classify it based on that information. However, if the decision cannot be made at that level, then the network goes to the next level consisting of another set of input features. There, it tries to classify the object using inputs at this level along with the information gained at the previous level. The authors believe that this approach will lead towards the successful establishment of human thinking and decision making in neural networks.<
>