Fuzzy Bayesian network research on knowledge reasoning model of food safety control in China

In order to take pre-crisis diagnosis, the safety warning, and definition of responsibility problems on the control of current food safety, a fuzzy Bayesian network of food safety risk knowledge reasoning model was established according to the data characteristics in the field of food safety control. Following the data research of the traceability system in the Bureau of Quality Supervision in a certain city in China, food safety risk related to index was extracted, whose value was defined by statistical methods. The sample data was thus achieved and the reasoning and diagnosis model of food safety control knowledge was set up by fuzzy Bayesian network algorithm based on the genetic algorithm. The application results show that fuzzy Bayesian network algorithm based on the genetic algorithm increased the computational complexity and running time due to the fuzzy math treatment, but the use of fuzzy logic can directly reflect the fuzzy random question reasoning and diagnosis on the possibility of high risk in certain step of the food production process. Compared with the general Bayesian network, the fuzzy Bayesian network has a higher accuracy of reasoning.