Semantic Search Engine, Thai Traditional Medicine Knowledge Graph Application

In this paper, a semantic information retrieval framework is presented to improve the precision of search results by concentrating on the context of concepts is presented. Instead of the keyword matching technique, the graph theory is used for leveraging the relationships and meanings of search terms. Information is represented as a knowledge graph for semantic searching performed by the inference engine and results are passed to the NLP technique for interpreting entity, relationship and property to their relevancy query. In the current system, we focused on healthcare system to helping the foundation recommendations of Thai Traditional Medicine. Once Thai Traditional Medicine contains a numerous of words which have the same meaning, the traditional keyword based cannot retrieve all relevant document. In our proposed semantic architecture, the query is performed an context-based understanding for retrieving all relevant document, therefore, it can retrieve the same number of documents for the queries which have the same meaning.