A fuzzy Petri net model for approximate reasoning and its application to medical diagnosis

The authors propose a fuzzy Petri net model to represent fuzzy production rules for a rule-based-system. In the proposed method, they use an agreement value (AV) where the AV is defined as a matching degree between the truth values of propositions and the possibility of the corresponding facts in order to give the consistency of rules with a result of reasoning. The model has not only the ability to represent knowledge but also systematic procedures for supporting fuzzy reasoning. To illustrate the proposed method the system has been applied to medical diagnosis and a satisfactory result was confirmed.<<ETX>>