Building nutritional meals at nursery school using fuzzy logic combined neural network and machine learning

Abstract

Nowadays, we have got many problems that we don’t have any correct and explicit ways to find a solution. AI (Artificial Intelligence) is an effective method which is used to find a solution for the problems. With AI of course, we don’t have an absolutely correct or true solution but only a solution that is nearly true. It means that we still have the gap between the true solution and the solution we get, but this can be used with good results. The task is to build nutritional meals at nursery schools where the input data and conditions for problem solving are not clear, so it is too difficult to have a correct and explicit algorithm to find the solution. We have a fuzzifier which performs the function of fuzzification which is a subjective valuation to transform measurement data into valuation of a subjective value, and then we use fuzzy logic to solve this problem. And we use neural network and machine learning to integrate this system to help us increase the correct rate and decrease running time. We have successfully built an algorithm and implemented it. Not only has this method helped to solve a particular problem but its use can also be extended to many other problems such as those about nutrition, schedule, etc.