Inference and improper knowledge detection on fuzzy rules with enhanced high-level fuzzy Petri nets

In the's paper, we propose a new version of Petri Nets - enhanced high-level fuzzy Petri net (EBLFPN) to represent fuzzy rules in knowledge bases. In the proposed model, fuzzy inference can be performed using a method similar to [3]. In addition, improper knowledge such as conflicts, redundancy, cycles, unnecessary conditions, dead ends, and unreachable goals, can be detected by first transforming fuzzy Petri nets into non-fuzzy Petri nets and then transforming the derived Petri nets into the reachability problems. At last, each type of improper knowledge is detected by solving a set of equations formulated from the corresponding reachability problem.