Consistency checking based on high level fuzzy Petri nets

Abstract

The problem of verifying the integrity of fuzzy knowledge bases is discussed. An approach to find potential inconsistencies in fuzzy rule based systems is described. The approach models the knowledge base as a high level fuzzy Petri net and uses the structural properties of the net for verification. Basic notions on approximate reasoning and high level fuzzy Petri nets are also given. The method used for consistency checking is briefly reviewed. Procedures for discovering potential inconsistencies at both local and global levels are described.<>