21st July 2008
14:00 - 19:00

Venue

University College London
Roberts Building G06, Sir Ambrose Fleming Lecture Theatre, Engineering Faculty
Roberts Building, Torrington Place, WC1E 7JE

This event, the third in ISKO UK's KOnnecting KOmmunities series, was an opportunity to learn more about SKOS (Simple Knowledge Organization System), the up-and-coming standard for publishing controlled vocabularies on the Web.

The topic proved to be enormously popular. There were 94 people present coming from government, publishing and broadcasting agencies, public service organisations, commercial companies and academia. Reassuringly many practitioners were present including subject matter experts, librarians, information architects, software developers etc.

We were fortunate in having speakers from the forefront of SKOS development and implementation who were able to draw a general picture of the role of knowledge organization systems in the Semantic Web development but also guide us through the intricate details of what it means to express controlled vocabularies in such a way so that they their semantics can be exploited in information integration and discovery.

A lively discussion enabled all to develop their own thinking on the opportunities to share and even generate new information through the combination of linked information resources and the vocabularies supporting them.

The afternoon represented a refreshing step forward in thinking through the more powerful application of controlled vocabularies and offered a valuable learning experience.

As always, we are very grateful to Conrad Taylor, co-ordinator of the BCS KIDMM (Knowledge, Information, Data and Metadata Management) community, for recording and photographing the event on our behalf.


Alistair Miles
[PDF] [MP3]
"The Simple Knowledge Organization System (SKOS) in the Context of Semantic Web Deployment"

Alistair's talk introduced SKOS, a soon-to-be-completed W3C standard for publishing thesauri, classification schemes and subject headings as linked data in the Web. It presented SKOS in the context of the W3C's Semantic Web Activity, and in particular the work of the W3C's Semantic Web Deployment Working Group where other specifications are being developed for publishing linked data on the Web, for embedding linked data in Web pages, and for managing Semantic Web vocabularies. Finally, this talk took a mildly inquisitive look at the value propositions for linked data in the Web, and how knowledge organisation systems might be deployed in the Web for better information discovery.


Antoine Isaac
[PDF] [MP3]
"On practical aspects of enhancing semantic interoperability using SKOS and KOS alignment"

In the first part of the presentation various SKOS features and the precise role of SKOS in relation to knowledge organization system (KOS) models were summarised. Antoine discussed some practical issues that have to be overcome when representing KOSs, such as thesauri and classifications, in SKOS. We learned how different features of SKOS correspond to the features of a traditional (and non-traditional) KOS, and how to see whether there are any key features missing. The second part of the presentation addressed SKOS's potential in expressing relationships between different vocabularies, and demonstrated how SKOS mapping links can be utilised to express several coexisting vocabularies. Various scenarios -- and projects -- were presented, which could benefit from these aspects. Concrete aspects of vocabulary alignment -- requirements, mapping techniques and evaluation problems -- were then introduced and discussed from a practical perspective.

Stella Dextre
Clarke

[PDF] [MP3]*
Leonard
Will
 
[PDF]
Nicolas
Cochard

[PDF]
"The BS8723 thesaurus data model and exchange format, and its relationship to SKOS"

The BS 8723 data model is closely allied to that of SKOS. The speakers described how it accommodates concepts, terms, relationships, arrays and node labels, and the especially challenging issue of compound non-preferred terms. They considered how all these are represented in XML, comparing this with the RDF approach adopted by SKOS. BS 8723 is soon to be adopted as an international standard, and the talk outlined the opportunities this could present for a closer tie-in with SKOS.

* All three talks are available as one recording.


Douglas
Tudhope
[PDF] [MP3]*




Ceri
Binding

[PDF]








"SKOS-based semantic web services: experiences from the STAR project"

The AHRC funded STAR project (Semantic Technologies for Archaeological Resources) has developed web services for knowledge organisation systems (KOS) represented in SKOS RDF format. The current service operates on a repository of English Heritage thesauri converted to SKOS format, containing terms and concepts familiar to those working within the archaeological domain. The service is based on a subset of the SWAD-Europe SKOS API, with extensions for concept expansion and provides facilities for search, concept browsing and semantic expansion. The talk described the services, demonstrated a client prototype browser application and touched on some of the issues in converting the thesauri to SKOS. Future possibilities for extending the SKOS API were discussed. For more details of the current services and to download the client, see http://hypermedia.research.glam.ac.uk/kos/terminology_services/.

* Both talks are available as one recording.


Bernard Vatant
[PDF] [MP3]
"Wondering about either SKOS or Web Ontology Language (OWL)? Use both!"

Using SKOS to represent vocabularies does not mean that an information system does not need OWL-based expressivity for some part of its knowledge base. Similarly, if a system is OWL-driven it does not mean that it does not need SKOS vocabularies. For some time, Mondeca, a semantic web company, has been deploying generic and robust architecture solutions where both OWL and SKOS components co-exist. Such an approach is based on a number of various customer use cases and requirements. Examples of this architecture were presented by Bernard including how SPARQL constructive queries can be used to index the publication files of a knowledge base, as an alternative to the difficult declaration of mapping from concepts in an OWL ontology (classes or instances) to concepts in a SKOS thesaurus.