Continuous learning and inference of individual probability of SARS-CoV-2 infection based on interaction data

Abstract This study presents a new approach to determine the likelihood of asymptomatic carriers SARS-CoV-2 virus by using interaction-based continuous learning and inference individual probability (CLIIP) for contagious ranking. is developed based on an directed graph (IDG), multi-layer bidirectional path tracking searching. The IDG determined appearance timeline spatial data that can adapt over time. Additionally, takes into consideration incubation period several features represent real-world circumstances, such as number present. After each update confirmed cases, model collects interaction infers person’s getting infected status surrounding people. CLIIP validated individualized SEIR simulate contagion process. Compared traditional contact tracing methods, our significantly reduces screening quarantine required search potential much 94%.