Diet Recommendation according to Kilocalories and People’s Tastes

Malnutrition and eating disorders are a latent problem in our society which are generated by an inadequate combination of foods either by lack of time, money, knowledge or a specialist who can help to properly manage food with the macronutrients necessary for good nutrition. In this research we present an architecture of diet recommendation using fuzzy logic and first-order logic, the research is divided into three phases: first, people’s data such as age, weight, height, physical activity level and gender were taken into account to obtain the required daily kilocalories using fuzzy logic; second, we considered as a knowledge base the menu plan for breakfast, mid-morning snack, lunch, mid-afternoon snack and dinner according to the tastes of the person for the first order logic; third, using a selection algorithm, a daily menu plan according to its kilocalories and the list of menus obtained with the first order logic are recommended. To validate the proposed architecture, Kaggle’s Cardiovascular Disease Detection dataset has been taken from which 500 people data have been taken for the research, the preferences of each person have been added to the dataset, finally the prototype recommends the diet for the 500 people according to the required kilocalories, the average kilocalories required are 1776 and the average kilocalories of the recommended menus are 1864, being the difference of 88 kilocalories, we conclude that our prototype based on the proposed architecture performs a proper recommendation.