Intelligent ontological agent for diabetic food recommendation
Chang-Shing Lee, Mei‐Hui Wang, Huan-Chung Li, Wenhui Chen · 2008
Diabetes is a chronic illness that food intake affects the bodypsilas needs and insulinpsilas ability to lower blood sugar. This paper proposes an intelligent agent, called the personal food recommendation agent, based on the ontology model for diabetic food recommendation. The agent can create a meal plan according to a personpsilas lifestyle and particular health needs. The required knowledge is stored in the ontology model predefined by domain experts. It contains the Taiwanese food ontology and a set of personal food ontology. The personal food recommendation agent includes the ontology creating mechanism, the personal ontology filter, the food fuzzy number creating mechanism, the fuzzy inference mechanism, and the real-time recommendation mechanism. It retrieves the personal ontology and meal records to recommend a personal meal plan based on the fuzzy inference mechanism. An experimental platform has been constructed to test the performance of the agent. The results indicate that the proposed method can work effectively and alleviate the effort of a registered dietician.