Towards Smart Healthcare Management Based on Knowledge Graph Technology

Abstract

With the improvement of people's living standards, people pay more and more attention to healthcare, in which a healthy diet plays an important role. Therefore, a scientific knowledge management method about healthy diet which can integrate heterogeneous information from different sources and formats is urgently needed to reduce the information gaps and increase the utilization ratio of information. In this paper, we propose a healthy diet knowledge graph construction model that promotes the development of healthcare management. The model mainly consists of three modules: named entity recognition, relation recognition and entity relevance computation, which are implemented with conditional random fields, support vector machine and decision tree algorithms respectively. These three modules obtain good performances with 91.7%, 99% and 87% F1 score on the datasets from three different websites. Based on the above results, we build a healthy diet knowledge graph by using ontology which contains food, symptom, population, and nutrient element entities as well as relations between food and entities mentioned above, so that people can use it for diet recommendations and other tasks.