Figure 2

COEUS architecture overview. Data integration connectors for CSV, XML, SQL and SPARQL enable the triplification of data into each seed’s semantic storage (1). Data are selected from each external resource to match specific ontology predicates, generating a rich knowledge base (2). Acquired data are available through COEUS' API, using Java methods, REST services, a SPARQL endpoint and the LinkedData view (3). The SPARQL endpoint enables the connection of multiple knowledge seeds through a transparent knowledge federation layer (4). With one or more seeds in place, the COEUS platform enables answering specific research questions or deploying multiple applications to web, desktop or mobile environments (5).