Integration of knowledge-based systems and neural networks: neuro-expert Petri net models and applications

This paper identifies and describes how knowledge based systems (KBSs), fuzzy logic and artificial neural networks (ANNs) can be integrated, and provides a novel (fuzzy) expert Petri net (EPN (FEPN)) based approach for the integration of KBSs and ANNs. A generic expert Petri net model of single neuron is presented, and a two-layer generic Petri net model for (fuzzy) neural networks and neural (fuzzy) expert Petri nets (NEPN, NFEPN)) that use this neuron model as a building block is described. The NEPN and NFEPN models can be used for representing a (fuzzy) knowledge base and (fuzzy) reasoning. Also, they can be utilized to develop ANN-like multilayered Petri net architectures of distributed hybrid intelligence having learning ability. Some application examples are illustrated for validation of the proposed models.