- From: Simon Spero <ses@unc.edu>
- Date: Thu, 6 Mar 2008 15:54:29 -0500
- To: "al@jku.at" <al@jku.at>
- Cc: "iperez@babel.ls.fi.upm.es" <iperez@babel.ls.fi.upm.es>, SKOS <public-esw-thes@w3.org>
Andy- Unfortunately SKOS currently can't support the kinds of inference you mention; it used to, but The Hierachical Relationship was been removed in the latest draft for reasons which aren't entirely obvious. Before these changes SKOS did provide this relationship in the form of the unqualifed skos:broader property. This corresponded directly to the Broader Term relationship which thesauri define in terms of document retrieval. The Hierachical Relationship is not necessarily valid when considered in terms of the underlying instances and classes of an OWL style ontology- it can happily cross BTP and BTG boundaries, etc. Currently the remaining vaue in SKOS comes from the label properties. These can be used with RDFS and OWL. See the discussions related to Issue-44 for more information. Simon Sent from my iPhone On Mar 5, 2008, at 11:25 AM, Andreas Langegger <al@jku.at> wrote: > > Dear SKOSsers, > > ....and I would like to add these: > > * When using Jena (for example), which reasoner can be used to make > inference on topics? > * How do I setup this such that a query like: > SELECT * WHERE { ?b a :Book; skos:subject c:Sports } also fetches > Books about Golf, Skiing, Wakeboarding, etc. > > Subsumption reasoning over classes is simple but effective for such > taxonomies like SKOS can be used for. And it should also work for > transitivity of properties that way. Because I think one of the > strange feelings about SKOS is the question, will I be worse off > when using SKOS instead of simple RDF Schema class hierarchies? At > least we are currently thinking about this because we would like to > use SKOS in a project. > > Thanks, > Andy > > On Mar 3, 2008, at 11:23 AM, Iván Pérez Domínguez wrote: > >> >> Hi, >> I've been reading about skos and I have a few questions that might >> make it to the FAQ page: >> * What's is SKOS? >> (after some nice description gives not much information) >> * No, seriously, what is SKOS? >> >> * When should I use it? >> >> * When should I not use it? >> >> * Can I use skos to model <specific problem>? >> >> I think this would be a nice starting point. >> >> Cheers, >> Iván. >> >> > > > ---------------------------------------------------------------------- > Dipl.-Ing.(FH) Andreas Langegger > Institute for Applied Knowledge Processing > Johannes Kepler University Linz > A-4040 Linz, Altenberger Straße 69 > http://www.langegger.at > > >
Received on Thursday, 6 March 2008 20:54:54 UTC