I
Even through SNOMED CT cannot be made available as a, say, RDF
endpoint, I think it is still useful to consider (non-
dereferenceable) URIs based on SNOMED CT concept identifiers for
annotation purposes in Semantic Web applications.
And as an IHTSDO rep, I agree with Olivier on the above point. I
honestly don't think one would want to "OWL-ize" SNOMED -- for various
reasons, it would yield a result that wouldn't resemble any kind of
OWL that anybody would ever come up with a priori.
To make it useful as OWL, among many other things, enormous numbers of
supplementary axioms would have to be added, and this would be an
"interpretive" activity that would change the meaning of the vocabulary.
But allowing mappings from OWL vocabs to SNOMED concepts (expressed as
URI's according to some standard formula), probably just as annotation
properties or using the SKOS vocab, might conceivably be worthwhile
for some people.
I've put before the SNOMED Ed Board in the past a request to publish a
standard formula for creating URI's out of SNOMED CT Identfiiers, but
the request has not been a high priority.
I can do it again. I'd probably just ask them to state that SNOMED
concepts could be represented "officially" as resources using a
formula like:
http://www.ihtsdo.org/sct# + the SCT identifier, which is a string of
numerals, typically about 10 characters long.
What exactly such a resource "is", is a question for....somebody else.
It certainly should NOT be mistaken for an rdfs:Class or an owl:Class.
I think it might be identified as an instance of a skos:Concept.
Anybody want to suggest a better formula? SCT identifiers are intended
to be stable, so while SNOMED does have versions, if there is a change
in a concept, it is supposed to get a new SCT ID; hence I'd just as
soon avoid the complications of indicating any versioning info in the
URI.
John
Olivier
Chimezie Ogbuji wrote:
On 7/17/08 4:22 PM, "eric neumann" <[EMAIL PROTECTED]> wrote:
As I had mentioned before, having all of NLM's MeSH and much of
UMLS available
as RDF/SKOS/OWL for researchers would be very well received by the
growing SW
community. I'd like to add this to the list of BioRDF activities
going
forwards.
Eric.
By UMLS, I'm assuming you mean the semantic network? Some time ago,
I wrote
a translation of the UMLS semantic network into an OWL document
that I would
be willing to contribute to the community, however, I wonder if
there are
licensing restrictions with NLM's MeSH and UMLS Semantic Network
that would
prevent the publication of a 'linkable' OWL ontology?
In generally, I'm asking because I had done something similar with
SNOMED-CT
and discovered (unless my understanding of the license is mistaken)
that
licensing restrictions of its kind essentially eliminates the
possibility of
SNOMED-CT hosted as 'open' linked data unless the hosting is done
by the
International Health Terminology Standards Development Organisation
(IHTSDO).
I wonder, how much of an impediment this is to the-reuse of current
terminology standards such as these in SW representational formats?
-- Chimezie
P Please consider the environment before printing this e-mail
Cleveland Clinic is ranked one of the top hospitals
in America by U.S. News & World Report (2007). Visit us online at http://www.clevelandclinic.org
for
a complete listing of our services, staff and
locations.
Confidentiality Note: This message is intended for use
only by the individual or entity to which it is addressed
and may contain information that is privileged,
confidential, and exempt from disclosure under applicable
law. If the reader of this message is not the intended
recipient or the employee or agent responsible for
delivering the message to the intended recipient, you are
hereby notified that any dissemination, distribution or
copying of this communication is strictly prohibited. If
you have received this communication in error, please
contact the sender immediately and destroy the material in
its entirety, whether electronic or hard copy. Thank you.
