A multilevel weighted reasoning algorithm using Interactive Weighted Petri Nets

This paper presents a multilevel weighted fuzzy reasoning algorithm for rule-based system based on interactive weighted fuzzy Petri nets (IWFPN). The proposed algorithm allows the local weights in the antecedents of the rule are represented by fuzzy numbers corresponding to the fuzzy linguistic terms of weights due to the interaction exists among attributes. Given a set of observed facts, the algorithm can perform fuzzy reasoning to evaluate the fuzzy sets of conclusions and their corresponding certainty factor values automatically. The proposed algorithm can allow the rule-based systems to perform fuzzy reasoning more intelligently.