Towards a Data-Driven Fuzzy-Geospatial Pandemic Modelling

The current Covid-19 worldwide outbreak has many lessons to be learned for the future. One area is the need for more powerful computational models that can support making better decisions in controlling future possible outbreaks, particularly when being made under uncertainties and imperfections. Motivated by the rich data being daily generated during the pandemic, our focus is on developing a data-driven model, not primarily relying on the mathematical epidemiology techniques. By investigating the implications of the current pandemic data, we propose a fuzzy-geospatial modelling approach, in which uncertainties and linguistic descriptions of data, some of which being geo-referenced, are handled by non-singleton fuzzy logic systems. In this paper, we outlining a conceptual model designed to be trained by the available pandemic worldwide data, and to be used to simulate the effect of an enforced controlling measure on the geographical extent of the infection. This can be considered as an uncertain decision support systems (UDSS) in controlling the pandemic in the future outbreaks.