Fuzzy Knowledge Learning via Adaptive Fuzzy Petri Net with Triangular Function Model

Since knowledge is vague and modified frequently in a expert system, this kind of rule-based systems are fuzzy and dynamic. It is very important to design a dynamic knowledge inference framework which is adjustable according to knowledge variation as human cognition and thinking. This paper presents an Adaptive Fuzzy Petri Net with Triangular function model (AFPNT). The fuzzy production rules in the rule-based system are modeled by AFPNT. Just as other fuzzy Petri net, AFPNT can be used for knowledge representation and reasoning. But AFPNT has one important advantage: it is suitable for vague and dynamic knowledge, i.e., the fuzzy model are adjustable by the data or the knowledge. Based on transition firing rules, a modification back propagation learning algorithm is developed for AFPNT to assure the convergence of the weights.