- From: Uschold, Michael F <michael.f.uschold@boeing.com>
- Date: Fri, 11 Nov 2005 07:35:42 -0800
- To: "Alan Rector" <rector@cs.man.ac.uk>
- Cc: "best-practice list" <public-swbp-wg@w3.org>, <chris.menzel@gmail.com>, <welty@us.ibm.com>, <dlm@ksl.stanford.edu>
Good example, Alan. ChrisM: lets see if we an add this to the examples in our narrative we are developing. CHrisW, and DeborahM: can you think of a nice example in the airplane/travel domain? Here's one I just thought of: "O1: hotel_reservation" "O1:flight_booking" O2:car_rental_booking" and "O2:plane_reservation" but no class that means booking or reservation. It could be a case of 'poor' ontology development, where there should have been the more abstract class. It may also arise when bringing two ontologies together, where in each, there was no need for the more abstract class. In either case, it involves creating a new class, and stating a subclass relationships. Mike -----Original Message----- From: Alan Rector [mailto:rector@cs.man.ac.uk] Sent: Saturday, November 05, 2005 5:40 PM To: Uschold, Michael F Cc: best-practice list Subject: Interoperability note Mike An important case we have come across frequently in mapping medical classes is where they are both subclasses of an interesting common superclass that is present in neither ontology, usually formed by deleting one or two restrictions from each. Our usual examples come from linking up radiology and surgery where we have several different versions of repair of bleeding aneurism by different methods but not common parent simply 'repair of bleeding aneurism' Closely tied to non-standard reasoning about least common subsumers and differences. Worth a chat Alan ----------------------- Alan Rector Professor of Medical Informatics Department of Computer Science University of Manchester Manchester M13 9PL, UK TEL +44 (0) 161 275 6188/6149 FAX +44 (0) 161 275 6204 www.cs.man.ac.uk/mig www.clinical-esciences.org www.co-ode.org
Received on Friday, 11 November 2005 15:38:10 UTC