Publishing and querying COVID-19 data as linked data

Today, tools for evaluating the current situation and performing analysis are gaining importance in order to help cope with the COVID-19 pandemic and prepare for possible future crises. In this context, many organizations have developed systems and projects to collect and publish global data on the impact of the virus, allowing monitoring of the COVID-19 pandemic. At the international level, COVID-19 data has been published on the web as open data to a certain extent. At the national level, the Ministry of Health has shared COVID-19 data in tables on a daily basis. In addition, COVID-19 data published by the Ministry of Health are organized into tables and shared as open data on the TURCOVID19 web page. These tables are available to everyone as statistical graphics and information on the page. It is significant and necessary to create semantic definitions of the data in such a large data collection using Semantic Web Technologies. Using COVID-19 data presented as open data, semantic definitions need to be created in order to extract new information and data from existing data. A full understanding of the COVID-19 data of the epidemic requires evaluating all data and mastering the results. Especially, making semantic definitions and mappings is the most important requirement for inferring new data at the schema and data level. Ontologies which are defined as "the conceptualization of concepts in a particular field" are developed for the creation of semantic definitions. The TURCOVID19 ontology was developed by evaluating existing ontologies to create semantic definitions of COVID-19 data. Developed TURCOVID19 Ontology and TURCOVID19 data were associated with Karma 2.2 software. Semantic descriptions of TURCOVID19 open data are published as RDF. Links are created between Silk Link Discovery Framework and entities in different existing Web data sources. These RDF links can be published with the original RDF COVID-19 dataset in the LOD cloud. Fuseki triple store is used for data storage and SPARQL endpoint. Using the SPARQL query interface, the RDF COVID-19 dataset was queryed and new information was extracted using semantic definitions and existing classes and relationships.