The Simple Knowledge Organization System (SKOS) is a core data model to publish vocabularies (thesauri, classification systems, terminologies, subject heading lists) on the web, as Linked Data. This can be especially useful for domains with a strong tradition in information science, such as libraries or museums. There, many high-quality vocabularies, like the Getty Art and Architecture Thesaurus (AAT), have been curated and maintained to help describing and finding resources. These can benefit advanced discovery scenarios, such as multilingual or semantic search, as currently happens for many collections in the Europeana.eu portal.
The creation of SKOS has accompanied a paradigm shift in the KOS community, from a term-based approach to concept-based approach. The traditional guidelines for creating and publishing vocabularies, like ISO 25964, are now moving ahead, standardizing finer-grained relationships between concepts (broader generic (BTG), broader partitive (BTP), and broader instantial (BTI)). This has resulted in new extensions to SKOS. However, these had so far not been provided with a complete set of formalized semantics, usable for advanced reasoning and reaping the benefits from such richer conceptual networks.
Together with Vladimir Alexiev (from Ontotext) and Jutta Lindenthal, both experts in the field and strongly involved in the modeling and publication of AAT, I have written a paper that seeks to address this gap. We have studied which formal axioms can be made about the new relationships, and applied them to the case of AAT, which has been recently published as Linked Data. It’s in line with a strong tradition of the Web & Media Lab, which has been researching the topic of converting knowledge organization systems to Semantic Web formats for well over a decade, especially for the cultural sector.
The paper is openly accessible via the website of Springer’s International Journal on Digital Libraries: http://link.springer.com/article/10.1007/s00799-015-0162-2