Epiwork Deliverable 3.5: Epidemic Data Ontology

We detail our work on evaluating and selecting a network of related ontologies for caracterising information relevant to the epidemiological domain. Instead of defning a new ontology from scratc, we propose a Network of Epidemiology-Related Ontologies (NERO), whic can be combined to form the core of semantic models in epidemic forecasting infrastructures. Departing from the metadata model of Epiwork's Epidemic Marketplace (EM), we evaluated existing proposals of ontologies for the epidemiology domain together with other ontologies that, despite having a diferent purpose, caracterise information frequently manipulated by epidemiologists and public health scientists. For the most part, ontologies in NERO are current candidates to the Open Biological and Biomedical Ontologies (OBO) initiative, a large community efort to establish a suite of reference ontologies for the biomedical feld. As part of the integration of Semantic Web tecnologies into the epidemiological domain, we expose how NERO ontologies can be explored to bring the EM into a more knowledge-oriented repository rather than a simple content database. As suc, we delineate the principles through whic semantic similarity, ontology matcing, and ontology extension can be used to further enhance text mining and other data processing activities in this infrastructure.