P11-02
Japanese Food Ontology Development
1Artificial Intelligence Center for Health and Biomedical Research (ArCHER), National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN)
2School of Medical and Dental Sciences, Tokyo Medial and Dental University
3Graduate School of Medicine, Kyoto University
4Graduate School of Science, Technology and Innovation, Kobe University
5National Cerebral and Cardiovascular Center
( * E-mail: higuchi@nibiohn.go.jp )
Nutrition is a key component of good health. However, many individuals may face challenges in obtaining sufficient nutrition due to factors such as allergies that affect their health. To conduct nutritional research that yields reliable and comparable results, it is necessary to use uniform terminology and accurate food descriptions. Consequently, there is a demand for a computer-readable Japanese food ontology that can precisely represent the characteristics and relationships of various foods.
The National Health and Nutrition Survey (NHNS) is a data collection initiative aimed at assessing the nutritional intake and lifestyle of the Japanese population, with the ultimate goal of improving their health. The NHNS involves physical measurements and blood tests from 3,412 households across 300 districts in Japan. The survey covers 1,630 Japanese food items, classified into three tiers: large categories (e.g., cereals, legumes, vegetables), medium categories (e.g., rice and its processed products, wheat and its processed products), and small categories (e.g., flour products, bread products).
We employed the Web Ontology Language (OWL) to describe the NHNS data, utilizing its hierarchical classification scheme and appropriate Uniform Resource Identifiers (URIs). This ontology is published as an alpha version of FGNHNS on BioPortal (https://bioportal.bioontology.org/ontologies/FGNHNS). Our future work includes enhancing the ontology by adding Wikidata information, linking with FoodOn, integrating with the Standard Tables of Food Composition in Japan, linking with the Agricultural Vocabulary System, and incorporating food allergy information.
In developing this ontology, we have been leveraging basic models such as Large Language Models (LLMs), and the environment surrounding LLMs has recently evolved further. In this poster presentation, we will introduce our current efforts based on the latest updates.