The SWAD-Europe Thesaurus Activity project is a SWAD-E sub-project aimed at showing how semantic web technology can be used by the KOS community for enhancing description of simple thesauri and classification schemes.
The focus of the demonstration was to show the end result of applying the developed models, techniques and systems using example data inside a web service application framework. This would show how the pieces connected together and while providing access to the core concepts, provides additional possibilities with standard semantic web tools.
The key outputs of the project are:
All data, software and documents were developed using open technical standards and open source.
Key goals of the project include:
These goals were approached by discussing with the community their issues and discovering use cases based on existing applications. This led to the development of the SKOS schema, with lots of influence from earlier work, with an appropriate modeling approach (concept based) and a gentle guide so that the path to using it was easy to achieve for those new to the technology.
The API and web services show how such systems can be made available and used with commodity technology (XML, SOAP and Web Services) and open source to leverage a high-level application. The resulting system as it was built with semantic web tools can then provide additional benefits such as RDF query, as the standardisation of the area continues.
The activity here continues in the mailing list hosted by the W3C as part of the SWBPD WG using community support tools such as Wiki to improve and extend this work beyond the end of the project.
The demonstrator illustrates the use of a simple RDF schema and some careful modeling to capture the core features of interest to a community, bringing it to the semantic web. The aggregation features of RDF allow different SKOS schemas to be manipulated together, and possible mapped at a later date, or by independent parties. The developing provenance tracking of semantic web tools can be used to enhance such mappings or aggregations by following who made such descriptions.
The modeling of the information in the individual thesauri also has been an important development such as the process of chosing the URIs for concepts and how and when to relate and group concepts with terms. This remains somewhat ongoing research as there are tradeoffs here; this is continuing in a different track of the SWBPD WG.
This demonstrator primarily shows the semantic web components: distributed data + core schema + local schemas + webness.