Food recommender system based on weighted ingredients, body mass index and allergies; using the Random Forest algorithm

Nowadays people live from day to day in a very hurried way and due to lack of time and knowledge they eat anything regardless of their health, allergies or level of food consumption. Knowing the standard of living of people and the food consumed is very important in this research. The objective is to design a system based on an intelligent algorithm that helps to recommend a list of ingredients that is adapted to each type of user based on the input data mentioned above. It is proposed to extract data on health and level of ingredient consumption by users through the Food Express application (Atriano Ponce, Saldaña Conde, Bello Garcia, Martinez Gorospe, & Sanchez Lucero, 2019), currently operating for the State of Tlaxcala. All this to generate a set of data that is analyzed by the Random Forest machine learning algorithm. Among the results, it is observed that the Random Forest algorithm has a precision of 97.7381% when making the decision to recommend or not an ingredient according to the input data of each user, while the ID3 algorithm has a precision of 94.0476 % and the Random Tree 87.5%. So, the Random Forest algorithm is more accurate for decision making.