Services for Large Scale Semantic Integration of Data

In recent years, there has been an international trend towards publishing open data and an attempt to comply with standards and good practices that make it easier to find, reuse and exploit open data. Linked Data is one such way of publishing structured data and thousands of such datasets have already been published from various domains. However, the semantic integration of data from these datasets at a large (global) scale has not yet been achieved, and this is perhaps one of the biggest challenges of computing today. As an example, suppose we would like to find and examine all digitally available data about Aristotle in the world of Linked Data. Even if one starts from DBpedia (the database derived by analysing Wikipedia), specifically from the URI “http://dbpedia.org/resource/Aristotle” it is not possible to retrieve all the available data because we should first find ALL equivalent URIs that are used to refer to Aristotle. In the world of Linked Data, equivalence is expressed with “owl:sameAs” relationships. However, since this relation is transitive, one should be aware of the contents of all LOD datasets (of which there are currently thousands) in order to compute the transitive closure of “owl:sameAs”, otherwise we would fail to find all equivalent URIs. Consequently, in order to find all URIs about Aristotle, which in turn would be the lever for retrieving all data about Aristotle, we have to index and enrich numerous datasets, through cross-dataset inference