W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > January 2007

Re: RDF resource name generation

From: Chimezie Ogbuji <ogbujic@bio.ri.ccf.org>
Date: Tue, 9 Jan 2007 18:58:59 -0500 (EST)
To: Alessio Carenini <carenini@gmail.com>
cc: public-semweb-lifesci@w3.org
Message-ID: <Pine.GSO.4.60.0701091842200.11424@joplin.bio.ri.ccf.org>

On Tue, 9 Jan 2007, Alessio Carenini wrote:

> Hello all,
> I'm developing a semantic interoperability system, trying to aggregate
> data coming from various HL7 v.3 documents in order to build instances
> of other HL7 v.3 documents. I've done a rough mapping of the HL7 v.3 MIF
> files into OWL ontologies and I'm finishing the algorithm to convert XML
> documents into instances of HL7 v.3 ontology.
> My problem is the generation of the resource names in RDF instances. If
> I use only blank nodes while generating triples, I won't be able to
> merge RDF graphs.
> Could someone point me to some document/article where this issue is
> discussed?

I've been working on a (related?) similar effort to map HL7 RIM XML 
documents to an ontology (so I'm very interested in this space).  With 
regards to your question, merging non-lean graphs (is that the proper 
term?) is mostly a matter of using terms that facilitate equivalence 
(functional properties - specifically), otherwise it becomes a much more 
difficult prospect.  You may find these links useful:

[1] http://esw.w3.org/topic/RdfSmushing
[2] http://esw.w3.org/topic/DeltaView

Chimezie Ogbuji
Lead Systems Analyst
Thoracic and Cardiovascular Surgery
Cleveland Clinic Foundation
9500 Euclid Avenue/ W26
Cleveland, Ohio 44195
Office: (216)444-8593
ogbujic@ccf.org
Received on Tuesday, 9 January 2007 23:59:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:16:37 GMT