The Personalized Traditional Medicine Recommendation System Using Ontology and Rule Inference Approach

The purpose of this research was to solve the complication associated with the recommendation of traditional herbal medicines, concerning the fact that an appropriate use of traditional herbal medicines entails contemplation of personal health information that include age, body temperature, pregnancy, lactation, chronic diseases, and medicines taken on a regular basis. Likewise, some traditional herbal medicines cannot be taken by patients with certain health conditions. Accordingly, this research proposed a system that provides recommendations of traditional herbal medicines according to each patient’s health information by applying an ontology-based knowledge representation technique that employs Web Ontology Language (OWL) to process and describe data in the ontology. Rules were expressed in the form of a rule language so as to enable the computer to infer and provide recommendations of traditional herbal medicines and their contraindications in a similar manner to a medical specialist. To test the efficiency of the proposed system in solving the complication and providing personalized recommendations, an experiment was conducted based on three scenarios: (1) the case of more than one diseases with different personal health information; (2) the case of more than one diseases with specified personal health information; and (3) the case of same disease with different personal health information. Upon assessment of the system efficiency by a medical specialist, the system was found to be capable of providing personalized recommendations of traditional herbal medicines and their contraindications in an efficient manner.