Analyzing Impact of Climate Variability on COVID-19 Outbreak: A Semantically-enhanced Theory-guided Data-driven Approach

ABSTRACT

With the intention of complementing the current worldwide actions to fight against novel coronavirus disease (COVID-19), substantial number of research works have been put forth during past few months so as to explore whether or how the various climatic factors influence the spread of this potentially fatal disease. However, because of uneven distribution as well as inadequate number of COVID tests, and also, due to lack of data transparency, these research findings are often found to be contradictory. In order to tackle such data inadequacy and uncertainty issues, in this work, we propose a theory-guided data-driven probabilistic framework with embedded technology of upgrading the impact analysis through incorporated climate domain semantics. Infusion of both the theoretical knowledge from epidemiology and the semantic knowledge from climatological domain helps the framework in better dealing with the uncertainty while appropriately capturing the pandemic characteristics of the disease. The effectiveness of our semantically-enhanced theory-guided data-driven approach is validated in terms of analyzing the causal influence as well as impact of climate variability on COVID-19 outbreak in several states of India. © 2021 ACM.