Learning and reasoning method using fuzzy coloured Petri Nets under uncertainty

Petri Nets have been widely used to model computer systems, manufacturing systems, robotics systems, knowledge-based systems, and other kinds of engineering applications, Further, to present complex real-world knowledge Fuzzy Petri Net Models have been proposed to perform fuzzy reasoning automatically. However, in the Petri Nets we have to represent all kinds of processes by separate subnets even though the process has the same behavior of other one, Real-world knowledge often contain many parts which are similar, but not identical. This means that the total Petri Nets becomes very large. Therefore, it becomes difficult to see the similarities and the differences among the individual subnets representing similar parts. The kind of problems may be annoying for a small system, and it may be catastrophic for the description of a large-scale system. To avoid this kind of problems we propose a learning and reasoning method using Fuzzy Coloured Petri Nets (FCPN) under uncertainty. On the other hand to correct the rules of knowledge-based system hand-built classifier and empirical learning method both based on domain cheery have been proposed as machine learning methods, where there is a significant gap between the knowledge-intensive approach in the former and the virtually knowledge-free approach in the letter. To resolve such problems simultaneously we propose a hybrid learning method which is built on the top of knowledge-based Fuzzy Coloured Petri Net and Genetic Algorithms. Since we may meet the relation among rules are ambiguous in practical applications, an algorithm is presented for checking the consistency of fuzzy knowledge base. via a learning process using Genetic Algorithm, Using this method, it keeps equally amount of information, while it decrease its net size, computing time and difficulty on explanation of network, and it can provide a structured representation in which die relationship among the rules in the knowledge base is easily understood. To verify the validity and the effectiveness of the proposed method, the system has been successfully applied to a medical diagnostic system.