Re: Mapping SKOS into BFO

On Thu, Apr 7, 2011 at 12:14 AM, Peter Ansell <ansell.peter@gmail.com>wrote:

>
> If SKOS:Concept was a child of BFO:GenericallyDependentContinuant, then you
> are restricting the published definition for SKOS:Concept. SKOS:Concept is
> deliberately defined subjectively as an open ended
> class that can be used to classify any unit of thought, whether
> it intimately relates to a single physical entity or otherwise.
>
> It may be more appropriate to define BFO: DependentContinuant as a subclass
> of SKOS:Concept.
>

BFO is committed to the existence of Universals, so that's not an issue

Concepts in KOS are intentional in nature; they are used to make statements
of what a  "document" is about- in this sense the concept is a unit of
thought.  For example, the hierarchical relationship (BT) in KOSs shows that
they are dealing with "Word and Subject".   There is usually an ontology
that can be induced from a KOS, but the process is rather subtle, especially
when the KOS has compound terms, is subdivided, etc.

The best currently available text  is Svenonious (2000).

Simon


Svenonius, Elaine (2000). The Intellectual Foundation of Information
Organization <http://www.netlibrary.com/AccessProduct.aspx?ProductId=39954>.
Cambridge, Mass.: MIT

Press. isbn: 0262194333 (hc : alk. paper).