Improving semantic interoperability of big data for epidemiological surveillance

Future disease outbreaks may spread faster and stronger than recent epidemics, such as Zika, Ebola and Influenza. The integration of multiple existing Early Warning Systems (EWS) is a requirement to support disease surveillance in combating infectious disease outbreaks. In this direction, numerous applications have been developed considering big data from diverse sources. However, big data potential can only be exploited if interoperability challenges are addressed. In this paper we discuss a semantic interoperability problem when using data exchanging standards for EWS integration. Particularly, we identify an issue regarding the distinction between the concepts of situation and event in the Emergency Data Exchange Language (EDXL), the OASIS set of standards for disaster management. To cope with interoperability issues we propose an ontology-driven situation-aware approach for the development and integration of EWS. The approach leverages on the OntoEmerge core ontology, in which we incorporate the clear distinction between the concepts of situation and event.