Ontology-Based Healthy Food Recommendation for the Elderly

The elderly population is particularly vulnerable to declining health, which often starts with reduced nutritional intake. As people age, their nutritional needs change, and inadequate intake can lead to chronic diseases such as heart disease, diabetes, hypertension and stroke. In Indonesia, the Nutritional Adequacy Ratio (NACR) focuses on the need for a balanced daily intake of nutrients including protein, fat, carbohydrates, fiber, vitamins, and minerals for the population aged 50-64 years. few studies have successfully developed food recommendations for the elderly. Several studies have successfully developed food recommender system applications for the elderly, However, this system is not for the elderly in Indonesia. Because Indonesian people have unique physical characteristics and food menus, and are different from other countries. This research addresses this problem by developing a system that is customized to the physical characteristic and food menu found in Indonesia. The system uses ontology to represent knowledge about healthy food choices and Semantic Web Rule Language (SWRL) to personalize recommendations based on individual health profiles and preferences. A Telegram chatbot serves as the user interface for the system, so that elderly people can easily access the recommendations. We evaluate the system using some metrics, such as Precision, Recall, and F-Score. The testing results show an F-Score of 0.933.