Re: Ontologies
thompson@zuben.ca.boeing.com (John Thompson)
Date: Wed, 7 Jun 1995 16:59:11 -0700
From: thompson@zuben.ca.boeing.com (John Thompson)
Message-id: <9506072359.AA00746@antares.ca.boeing.com>
To: cg@cs.umn.edu, srkb@cs.umbc.edu
Subject: Re: Ontologies
X-Sun-Charset: US-ASCII
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
Toward the end of his long and interesting letter Patrick Cassidy
wrote:
>... Even with
>the top 2000 concepts agreed upon, there will still be enormous
>scope for proprietary details and variations in the programs that
>will use the ontology.
> To be really standard, an ontology should also include
>some level of agreement on semantic relations (slots) that are
>used in defining terms. Without some commonality in relational
>links, it will be hard to verify that the meanings of the
>individual nodes in a semantic network have the same
>significance to all systems.
Yes, and the semantic relations (slots) should be organized as part of
the ontology itself. In the Cyc ontology, if you go down about 6 or 7
levels you come to the collections Predicate and Slot (binary
predicate), and under those (as instances) are the actual predicates,
arranged in their own ontology. I don't know if being that deep in the
ontology would qualify them for inclusion in the "top 2000" list, but I
feel that they are the most important and valuable part of the
ontology.
Coming to some agreement on an ontolgy of standard predicates should be
at least as high priority as coming to agreement on general
categories. An extremely valuable paper that the Cyc people could
write would be a study of which of their predicates are used most
frequently throughout their knowledge base(s), along with a discussion
of the design, meaning, and use of each of them. Some of Cyc's most
basic predicates as of 5 years ago were discussed throughout section
3.3 (Details of Inferencing in CycL) in Lenat & Guha's book _Building
Large Knowledge-Based Systems_.
John A. Thompson
The Boeing Company
thompson@zuben.ca.boeing.com