Heuristic approach to topology generation for knowledge-based fuzzy Petri nets

Using fuzzy Petri nets as expert knowledge representation is very effective for implementing a fuzzy reasoning system. However, these systems cannot generate new rules voluntarily. To avoid this kind of problem, we propose an algorithm to improve the initial domain theory of an expert. The proposed method generates a new node where the node is evaluated in the network. The result can be more general knowledge than the initial knowledge. To verify the validity and effectiveness of the proposed method, the system is applied to a medical diagnostic system.