RE: Ontology Mapping

Bill,
Lewis departed for vacation at lunch time and won't be back till thursday.

So as not to leave you haning for an answer till then...

As taken from http://codip.grci.com/Articulation.daml the comment field for
Match.
"A specialization of ObjectProperty that matches its non-empty set of
subjects to its non-empty set of objects."  
The mapping is asymmetric, from the subject ontology to the object ontology.
as described on http://codip.grci.com/Tools/ArtiServicePage.html
More details are discussed on the Service Page.

Seems similar to DCG's but with a Match the specification of the translation
between
the subject(set) and object(set) is optional.  

Pat
__________________________________________
Patrick Emery 
AT&T Government Solutions 
1900 Gallows Rd.       Voice (703)506-5259
Vienna, Va 22182       Fax   (703)556-4261

-----Original Message-----
From: Bill Andersen [mailto:andersen@ontologyworks.com]
Sent: Friday, August 09, 2002 12:57 PM
To: Hart, Lewis; Hector Ceballos
Cc: RDF Logic
Subject: Re: Ontology Mapping



On 8/9/02 7:27, "Hart, Lewis" <lhart@grci.com> wrote:

> We have implemented one-to-many mapping between two ontologies by using a
> third ontology, an Articulation [1]. It uses a Match class with
multi-valued
> subject and objects, for example:
> 
> <art:Match rdf:ID='M1'>
> <subject rdf:reference = &o1;name>
> <object rdf:reference = &o2;first_name>
> <object rdf:reference = &o2;last_name>
> </art:Match>

Hi, Lewis..

Could you please tell me what the semantics of the above are?  Looks to me
doing something similar to a DCG rule in Prolog.  Thanks.

  .bill