Re: SUO: RE: Architectures for Intelligent Systems
> Tim King a écrit :
>
> I am not sure that I see the following to be truly rigorous:
>
> > Non basic attributes/properties have the form of binary relation
> > where one argument (say the first) is an instance of the concept
> > being defined and the other argument is an existentially quantified
> > occurrence of another concept.
> >
> > A car has a 'power source' which is an 'engine'.
> >
> > Thus, the relation itself must be "conceptualised" with respect to
> > it's "function", "role" or "place" within the defined concept.
> > It is not enough to say "a car 'has' an engine".
> >
>
> You have used the word "has" twice and claimed some difference.
Yes! bad phrasing, it means: ('power source of' 'car' 'engine')
but I screwed the translation to plain english.
> You seem to imply that "power source" and "engine" are different.
Yes! "power source of" is a predicate and "engine" is a concept.
> Is this all the triadic stuff that Jon expounds. In which
> case, perhaps I can see that expressing the situation in natural language is a problem.
> Ultimately, people do say "My car has an engine." Surely this must have a representation too?
Yes! "My car has an engine." = ('some kinda stuff of' 'car' 'engine')
They know that there must be some use for it, but even if they
know what (a power source) they don't care to mention it.
> Even if one that is far from being specific in the ways suggested.
That raise the interesting question of subsumption among predicates.
'some kinda stuff of' is less specific than and subsumes 'power source of'
That should not prevent the broker to figure out that an ontology
that is less specific about some property of a concept is nevertheless
referring to the same concept, it has to rely on other details.
But the FCA concept closure is quite handy for that it will map back
the incomplete set of properties to the local proper concept that
best match the incomplete set provided this one is specific enough.
This is usefull to identify a concept among user input chatter but less
so when querying the source broker for the "meaning" of a concept because
then, it must respond with the *full* set of properties for the concept
and there will likely be enough redundancies to sort out the subsumptions
from the true divergences about the concept comprehension.
Thank you for your remarks.
-- Jean-Luc Delatre
--------------------------------------------------------------
"In order to make an apple pie from scratch, you must first
create the universe." - Carl Sagan, Cosmos
--------------------------------------------------------------
http://perso.club-internet.fr/jld/ -- GSM: +33 6 11 24 06 29