Food Menu Recommendations using Content-based Filtering Method

Abstract

A nutritionally imbalanced body poses significant health risks. Achieving nutritional balance and maintaining a healthy diet are therefore essential. One of the major challenges in this regard arises when a user's dietary preferences conflict with their health conditions. To address this issue, it is necessary to develop a system that not only identifies nutritional needs and assesses their adequacy but also takes into account users’ medical histories. This study aims to develop a food menu recommender system that integrates nutritional requirements, user preferences, and disease history. Content-based filtering was employed to measure the similarity between user profiles and the available dataset. A weighting model was incorporated to accommodate food restrictions associated with specific health conditions. The dataset used in this study was obtained from the Center for Nutrition Research and Development and Food, Ministry of Health of the Republic of Indonesia. The research process includes data collection, pre-processing, user profile identification, and the generation of food recommendations. Additionally, this study explores the integration of user preferences and disease records using a binary vector to improve recommendation accuracy. The findings indicate that the developed system successfully provides food recommendations that are not only calorically adequate and nutritionally balanced, but also consistent with the recommended macronutrient distribution for protein, fat, and carbohydrates.