Personalization Nutrition Recommendation System Based On Ontology

With the improving of people's standard of living, how to eat healthily becomes the focus of what people think of. Nowadays, there has been a lack of dietitians and the computer nutrition expert is popular for its convenience and high ability of calculating. Most of the nutrition expert systems are special for medical treatment or children, and they care whether the ten or more nutrition elements are balance most but ignore the preferring of the persons. What worse is that they are mostly for fee or stand-alone software but not web software, which is not prone to popularization.We present a new approach to the effective development of menu recommendation and construction systems that allow to automatically construct a menu that is strongly tailored to the individual requirements and food preferences of a client. Personalization Nutrition Recommendation System Based on Ontology (PNRSBO) uses an improved collaborative filtering skill and the context skill to compute that how much a person likes a kind of food. The degree of a client's liking of different food lies on the sensory similar of food, the similar properties of persons and food, and the people's effect on each other. Clients modify their menus which are recommended by our system is allowed and the system will learn from these which food clients like more. We use OWL (Web Ontology Language), which can express semantic information understood by computers, to save and transfer knowledge, and help computers to learn and do reasoning.We also bring forward a new menu-generating method, using Graph. And we think out a way to realize a computer food-therapeutic system with the jena reasoning API.Finally we have a look at the future that we can open up a new avenue towards building nutrition expert system based on SOA. With the skill of OWL-S/WSDL, our system would have access to the heterogeneous databases with different structures and in different plates, which would be a shared nutrition recommendation system in Semantic Web.