Hello again Yongchun --

[Yongchun]
Drew has a clear answer for that based on "rules", but what is the proper engine that can understand this rule? Since I do not "familiar" with Adrian's "Internet Business Logic" rule system. I'd like to get the answer from you, Adrian, to see the result. Please try and tell us.

[Adrian]
Here is one way to merge your ontologies (as you described, below) using the Internet Business Logic rule system.

There is just one rule.  It has the form premises, underline, conclusion:

  in Ontology1 a subclass of some-class with some-property equal some-value is named some-name1
  in Ontology2 some-class1 is a subclass of some-class2 and is named some-name2
  in a merged Ontology we wish to identify that-name1 with that-name2 and call it some-name
  -----------------------------------------------------------------------------------------------
  the merged Ontology contains a class that-name based on the classes that-name1 and that-name2

There are tables for each of the three premises (not shown here).  You can see the tables and run this example by pointing Netscape 7 or Mozilla to www.reengineeringllc.com and selecting MergeOntologies1 .  When you run the rule, you can get the following answer

  the merged Ontology contains a class Men based on the classes 1:Men and 2:Men

You can also get this explanation:

  in Ontology1 a subclass of Humans with Gender equal Male is named 1:Men
  in Ontology2 Males is a subclass of Humans and is named 2:Men
  in a merged Ontology we wish to identify 1:Men with 2:Men and call it Men
  -----------------------------------------------------------------------------
  the merged Ontology contains a class Men based on the classes 1:Men and 2:Men

Viewed this way, the example turns out to be quite simple.  Of course, further rules are needed to use the above with actual data (such as  Evan and Drew in 1:Men ,  Adrian in 2:Men) that the two ontologies describe.

A more interesting example that you can also run in Internet Business Logic is called SemanticResolution1, in which it is necessary to walk an is-a hierarchy of names in order to bridge the ontologies of a retailer and a manufacturer who wish to do business.

Thanks in advance for comments.                 Cheers,  -- Adrian

                                            INTERNET BUSINESS LOGIC

                                              www.reengineeringllc.com

Dr. Adrian Walker
Reengineering LLC
PO Box 1412
Bristol
CT 06011-1412 USA

Phone: USA 860 583 9677
Cell:    USA  860 830 2085
Fax:    USA  860 314 1029


At 02:42 PM 12/4/03 -0500, you wrote:
Hi, Adrian, Drew
 
Thank you very much for your clear answer.
 
Drew has a clear answer for that based on "rules", but what is the proper engine that can understand this rule? Since I do not "familiar" with Adrian's "Internet Business Logic" rule system. I'd like to get the answer from you, Adrian, to see the result. Please try and tell us.
 
Thank you :)
 
Yongchun
----- Original Message -----
From: Adrian Walker
To: Yongchun Gao
Cc:
Sent: Thursday, December 04, 2003 2:29 PM
Subject: Fwd: Re: relation between ontology

Hello Again Yongchun Gao --

Please let me know if you would like to try Drew's suggestion, below, in
our Internet Business Logic rules system.  (The system is live, online at
www.reengineeringllc.com )

If that is not high on your list of things to do, I'll write it and add it
to our Semantic Web demos.

                                 Cheers,  -- Adrian





                                            INTERNET BUSINESS LOGIC

                                              www.reengineeringllc.com

Dr. Adrian Walker
Reengineering LLC
PO Box 1412
Bristol
CT 06011-1412 USA

Phone: USA 860 583 9677
Cell:    USA  860 830 2085
Fax:    USA  860 314 1029









>X-Originating-IP: [18.29.1.71]
>X-Original-To: www-rdf-logic@frink.w3.org
>Delivered-To: www-rdf-logic@frink.w3.org
>Delivered-To: www-rdf-logic@w3.org
>Date: Thu, 4 Dec 2003 13:54:58 -0500 (EST)
>From: Drew McDermott <drew.mcdermott@yale.edu>
>To: www-rdf-logic@w3.org
>X-YaleITSMailFilter: Version 1.1d (attachment(s) not renamed)
>Subject: Re: relation between ontology
>X-Archived-At:
>http://www.w3.org/mid/200312041854.hB4Isvu02058@pantheon-po04.its.yale.edu
>Resent-From: www-rdf-logic@w3.org
>X-Mailing-List: <www-rdf-logic@w3.org> archive/latest/3991
>X-Loop: www-rdf-logic@w3.org
>Sender: www-rdf-logic-request@w3.org
>Resent-Sender: www-rdf-logic-request@w3.org
>List-Id: <www-rdf-logic.w3.org>
>List-Help: <http://www.w3.org/Mail/>
>List-Unsubscribe: <mailto:www-rdf-logic-request@w3.org?subject=unsubscribe>
>Resent-Date: Thu,  4 Dec 2003 13:55:00 -0500 (EST)
>
>
>
>    [Yongchun Gao]
>    Can OWL "unite" all the ontologies? Or OWL is just a way to represent =
>    ontology on the Web?
>
>The latter.
>
>    In OWL, there are subClassOf, subPropertyOf, etc. From first sight, it =
>    seems that OWL can unite at least some ontology. But giving an simple =
>    example.
>
>    Suppose someone developed an ontology by OWL, in which "humans" is a =
>    class and has "hasGender" as a property (value=3Dmale/female). A man =
>    could be an instance of "humans" which "hasGender" of "male". It can =
>    work well.
>
>    Suppose anther expert developed an ontology by OWL too, in which =
>    "humans" and "animals" are classes, but "females" and "males" are =
>    classes too (can be attached to both "animals" and "humans"), and "men" =
>    is just two subclass of both "humans" and "males". It may work too.
>
>    But the problem here is HOW to unite these two different OWL files which =
>    tell the same ontology?
>
>Import them into a new ontology, using namespace prefixes to keep
>'male' in the first ontology a distinct symbol from 'male' in the
>second.  The new ontology (which we call a _merged ontology_) then has
>a property T1:hasGender and an individual T1:Male that is a possible
>value of T1:hasGender, and it also has a class T2:Male.
>
>Add an axiom, or "rule," that says
>
>    forall (x) ((x member T1:humans) & (x T1:hasGender T1:Male)
>                <->
>                (x member T2:humans) & (x member T2:Male))
>
>Make sure you have an inference engine that can handle these rules.
>Many of them handle equivalences by expressing them as a conjunction
>of implications, so you may have to break this into two rules.
>
>--
>                                              -- Drew McDermott
>                                                 Yale University CS Dept.