Dynamic Adaptive Fuzzy Petri Nets for Knowledge Representation and Reasoning

Abstract

Although a promising tool for knowledge representation and reasoning, fuzzy Petri nets (FPNs) still suffer from some deficiencies. First, the parameters in current FPN models, such as weight, threshold, and certainty factor do not accurately represent increasingly complex knowledge-based expert systems and do not capture the dynamic nature of fuzzy knowledge. Second, the fuzzy rules of most existing knowledge inference frameworks are static and cannot be adjusted dynamically according to variations of antecedent propositions. To address these problems, we present a new type of FPN model, dynamic adaptive fuzzy Petri nets, for knowledge representation and reasoning. We also propose a max-algebra based parallel reasoning algorithm so that the reasoning process can be implemented automatically. As illustrated by a numerical example, the proposed model can well represent the experts' diverse experience and can implement the knowledge reasoning dynamically.