Semantic Interoperability Toolkit for Data Marketplaces

Due to the countless number of data sources, an integration issue emerges, as the same data often have several representations. Therefore, this paper suggests a Natural Language Processing (NLP) based approach for the so-called Semantic Interoperability Toolkit. It comprises two components - the Semantic Matchmaking solution and the Semantic Engine - allowing users to identify similarities between concepts and manage queries against them respectively, regardless of their original representations. The solution is at use on the AI4PublicPolicy's Data Marketplace for supporting the exploitation of a wide range of data sets.