Re: Equality and subclass axioms

pat hayes wrote:
> 
> 4. I think that semantic search engines will need to be able to
> integrate ontologies on the fly...
> Ans: Well, I agree this would be desirable, but if that 'integration'
> is going to involve finding a logically coherent consensus among
> potentially disagreeing agents, then  God alone knows how to do it,
> and even He might have trouble.

However, humans seem to do this kind of integration all the time. We
have all seen conflicting viewpoints on the election mess in Florida,
and most of us have come to a conclusion (one way or another) about the
situation. How we do this depends greatly on our prior beliefs (which
sources we trust, consistency with other things we believe we know to be
true, etc.). I think that we will need some formal notion of belief
systems (a series of rules that order and prioritize potentially
conflicting information) to make this kind of integration work. Sample
rules might be "I believe that person X is married to person Y only if
both X and Y say that X is married to Y", "I believe the Washington Post
more than I believe the National Enquirer" or "I believe person X is a
faculty member if an acredited university says that X is a faculty
member."

To reduce the effort required by an individual user, we could make these
systems sharable, so that at some point I could say "I subscribe to the
Republican Party belief system" or "I subscribe to the Democratic Party
belief system" when I ask the question "Who won the U.S. presidential
election?" and depending on which belief system I selected, I'd get a
different answer.

Jeff

Received on Tuesday, 28 November 2000 11:46:39 UTC