Providing a Smart Food Recommender System Based on People's Characteristics And Health Level Using Multilayer Perceptron Neural Network Classification

Today, the variety of food has increased in different societies, and one of the important things to increase the quality of life of the people of that society is to choose food according to the characteristics of those people. By considering all the characteristics of consumers together, a more suitable choice for your food can be considered. In any society, eating habits and its choice depend on various factors, some of these factors include climate characteristics, ethnic and religious characteristics, community culture, abundance of food, population, genetic characteristics, health of people, etc. . Many recommender systems have been developed for this purpose, most of which have not considered all food selection indicators based on user characteristics. Some of the presented techniques are not very flexible and some consider only one of the user's characteristics. In this article, we will examine the different methods of intelligent systems that are proposed to users using artificial intelligence and machine learning algorithms, and the strengths and weaknesses of each, so that we can respond to this challenge. Considering which user characteristics can be effective in the success rate of the recommender system. Finally, we present an intelligent food recommendation system method based on neural network classification based on people's BMI test. The data used is the data set of the European Union Statistical Office, which includes the BMI test results of people from different countries. The desired system shows an accuracy of ۹۹.۸۰%, which is an acceptable level of accuracy.

Keywords:

artificial intelligence , food recommender , health food , food Recommender Systems.