Interactive Visual Analysis of COVID-19 Epidemic Situation Using Geographic Knowledge Graph

Abstract

After the outbreak of coronavirus disease 2019(COVID-19), there are large amounts of data related to spatiotemporal information. It is difficult for the epidemic analysis with geographic spatiotemporal model to take the data of character relationship and events into consideration.Therefore, a method of using geographic knowledge graph and interactive visualization to analyze the epidemic situation of COVID-19 is proposed. First, the entity and relationship types of patients are defined, the event semantic model and event relationship classification are proposed, and according to different types of data, knowledge extraction and knowledge representation methods are designed, and the knowledge graph of patients' spatiotemporal information is constructed.Then, from the macro and micro level of the task of epidemic situation control, an analysis frame combining semantic network and visual analysis model is proposed. Finally, an experimental analysis system is built, which uses the data of confirmed patients from COVID-19 to carry out the experiment on the situation analysis of COVID-19 through multi-view collaborative interaction analysis such as map distribution visualization, knowledge graph visualization and trajectory visualization. The experiment has proved that it can provide a new way to analyze the epidemic situation from the aspects of real-time situation monitoring, patient relationship analysis, high-risk population analysis and regional control situation analysis.