Re: Cross-ontologies reasoning

Hi, Jim...

On Dec 29, 2003, at 1350, Jim Hendler wrote:

> Bill, I disagree, I'd actually claim that in the real world virtually 
> all of this stuff we're doing is heuristic at best, so I'm less 
> convinced that these techniques wouldn't work.  That said, I don't see 
> why partial matchings have any less value than total matching.

Perhaps we had a disagreement as to what the nature of cross-ontology 
mapping is.  When you mentioned the use of heuristic techniques, I had 
imagined that you meant something like:

   (x)[o1.bachelor(x) <-> o2.bachelor(x)]

by virtue of the fact that the symbol names were the same or some such. 
  But of course such a mapping is of limited utility unless one takes 
into account the constellation of axioms surrounding the respective 
terms in the respective ontologies.

Say in ontology o1 you have

   (x)[bachelor(x) <-> (male(x) & ~married(x))]

and in ontology o2 you have

   (x)[bachelor(x) <-> (male(x) & ~married(x) & age(x,y) & y<60)]

But this only works in the o2->o1 direction and then only if there is 
agreement on what 'male' and 'married' mean in each.  So, Rush Limbaugh 
notwithstanding, you don't really get to pick which axioms you don't 
care about because you don't know a priori which axioms the 
*applications* attached to the ontologies you wish to merge do care 
about and to what degree they care about them (e.g., can inconsistency 
be tolerated as in a content management system like Semagix).

To do so would imply an intimate knowledge of the intended semantics of 
both applications which is exactly what this whole semantic web 
business is trying to avoid - one simply takes the ontologies at face 
value and everything works out happily.

I do agree with you that human-in-the-loop will be required for the 
foreseeable future and I am in fact *not* pessimistic at all about the 
possibilities of ontology merging.  Quite the contrary.  I am very 
optimistic.  What I am pessimistic about is that logically weak and/or 
incomplete heuristic techniques will be of any use in mission critical 
applications.  I might use them to buy a new CD on the web but I would 
not use them to fly my plane, control my nuke plant, or monitor a heart 
patient.

>> Question: What does it mean "[to map] instances against multiple 
>> ontologies"?
>
> (I sent this to a couple people one-on-one, but since I was asked on 
> the list) --
>
> what I had in mind is something more like the fact that on the SW I 
> can say something like
>
> :Photo1 foaf:depicts :MyDog.
> :MyDog a cyc:Dog,  Petshops:Dog;
>  cyc:skinColor  colors:brown;
>  cyc:Breed Petshops:Labrador.
>
> From the above (which cam be easily created in RDF from various markup 
> tools floating around - cf http://owl.mindswap.org/images/) I 
> postulate a tool should be able to abduce some useful information not 
> just about MyDog, but about how cyc and petshops definitions of "dog" 
> are related -- note also that this would work if Petshops used some 
> other URIref than "Dog" -- canine, celeb, perro, etc...
>  There are several people (perhaps you folks included) working on 
> mapping database schemas/instances to each other and using that for 
> integration, but most of the ones I know map multiple databases to a 
> single ontology (or a set of disaparate, but unlinked ones), what I 
> haven't seen is a lot that is looking at instances like the above 
> which bring together info from multiple ontologies in interesting ways

That's a pretty cool idea.  Especially when a supplied ontology is 
weakly axiomatized, one can learn a lot about the intended meanings of 
the terms by examining instances, not only of objects, but also of 
relations, that fall under such an ontology in an application to which 
the ontology has been put.

Happy New Year!

   .bill

--
Bill Andersen (andersen@ontologyworks.com)
Chief Scientist
Ontology Works, Inc. (www.ontologyworks.com)
1132 Annapolis Road, Suite 104
Odenton, Maryland 21113
United States
Office: 410-674-7600
Mobile: 443-858-6444

Received on Monday, 29 December 2003 14:37:13 UTC