Proposal of COVID-19 infection risk ontology focusing on situations of activities and spaces

Abstract

To prevent the spread of the COVID-19, the Japanese government has indicated situations that may cause infections, as "Three Cs" and "Five scenes." To calculate the risk of individual behaviors based on the simplified rules, formal definition of such situations are needed, together with the behavioral and spatial information of people. In this study, we propose an ontology that can infer the degree of Three Cs in peoples' behavior from the interview of infected patients. The inference can be used to help the manual survey by public health authorities, and analysis of potential risks hidden in the actual tempo-spatial behaviors of people. This paper also discusses possible applications of the proposed ontology.