A Graph-Based Approach Towards Risk Alerting for COVID-19 Spread

Abstract

With the spiraling pandemic of the Coronavirus Disease 2019 (COVID-19), it has becoming inherently important to disseminate accurate and timely information about the disease. Due to the ubiquity of Internet connectivity and smart devices, social sensing is emerging as a dynamic sensing paradigm to collect real-time contacts between both people and places. For example, we can rely on the Bluetooth signals that smartphones can both send out and receive to collect the real-time user contacts data. Based on the contacts data, in this paper, we investigate to propose an efficient approach to calculate the risk level of each person to have COVID-19. It can help pinpoint the people who need to be isolated. (1) We model the real-time contact data between people as a straming graph, which is a constantly growing sequence of edges. (2) We provide a risk alerting model to find the people who came in contact with someone having COVID-19. (3) In addition, we design efficient algorithms to calculate the risk level of each person and update the levels in real time. (4) Extensive experiments verify the effectiveness and efficiency of our approach.