[Sigia-l] Ontologies vs. DM-Even Longer-Sorry!

[lisa colvin]

Hi Nuno,

Sorry for the long post. 

>In OO Modeling objects (instances) do not
>really exist explicitly but are spurred when the
interpretation of it
>takes on. 

Similarly, in cyc, instances are instantiated when the
need arises. (For example, when the inference engine
is asked to prove a fact or inference.)


>Suppose X and Y are classes, and pA is some
association (predicate) 
>that
>relates class X to Y (X pA Y, or pA(X, Y)).  If when
interpreting the
>model I find the object [X] of class X and [Y] of
class Y, then >pA([X],
>[Y]) may or not be true right? But again suppose when
interpreting the
>predicate I find that pA([X], [Y]) = True, in OOA
[pA([X], [Y])] is
>actually an instance of pA(X, Y) (in other words it
links object [A] >to
>[Y], so a Link). 

Hmmm. Well, maybe under-the-covers (in the inference
engine), there is instantiation of predicates.
However, to the naïve ontologist! there is no
instantiation of predicates, only of instances. This
may be due to the purpose of instantiation. Typically
I’m asking the system to prove the existence of an
individual instance. Thus, the system hypothesizes the
existence of this individual instance and through
lookup, backward- and forward-chaining rules, etc.
sees if the existence of the instance is logically
compatible with the rest of the rules/facts in the
ontology.


>I see. To demonstrate why multiple inheritance is
most often not >needed
>let me try to provide a sample. 

<snip>

>Human
>  Person
>    Employee
>      Parent
>    Parent
>      Employee
>Organization
>	Employer
  
>Humm, but I can't have that in OOA and I suspect that
neither in
>Ontologies, so let's flatten the model.

Right! This model is bad because the Parent-Employee
relationship is cyclical.

I probably wouldn’t create classes of individuals who
are defined strictly on a relationship. So, “parent”
would be a relationship between two individuals and
“employee” would be the relationship between an
individual and an organization. (“employer” could be
defined as the inverse, but it’s not necessary).

>Human
>  Person
>    Employee -> Parent & Employee
>    Parent -> Parent & Employee

Right! 

This model has bad logic.

Also, it’s not as useful to conflate these two classes
as one couldn’t be a childless employee or unemployed
parent.

Further, as you said later, the definition of Employee
is separate from Parent which may fluctuate from
domain to domain…

>there is definitely a relationship between a Parent
and Employee and
>Person but not necessarily a [is of] kind of
relationship. The kind of
>relationship that exists is more of a Role.

Right, or, as I would prefer, as a relationship.

>my
>point is that experience has shown that multiple
inheritances it is 
>only
>needed in very few extreme cases. I understand that
in linguistics it
>may be needed, but IMO that is only because language
we try to take
>language out of the context (in other words we try to
normalize 
>language
>according by using partial functions). 

I see your point. Multiple inheritance systems are
tricky and difficult to implement.  However, sometimes
it is necessary in order to make an ontology more
manageable. This is where the faceted approach to
ontology development comes in.

>This is interesting. Do you use some fuzzy scheme to
quantify the 
>degree
>of truthfulness of a predicate? What kind of
quantification scheme is
>used?

When I was at cyc we didn’t have fuzzy quantification,
only existential and universal. However, that was 4
years ago, so maybe it’s changed.

>Well this is all interesting indeed, but I'm more
interested in >playing
>the pipe of pure technology and pragmatics at the
moment. I've read in
>your other posts that you have the role of IA ;) in a
number of web
>projects.

What is in a name? Actually there is another “lisa
colvin”! She is an IA in Los Angeles. I have been
working mostly as an ontologist in the SF Bay Area.
This is very confusing. I found out when we both
applied for the same job! Pretty funny.  So,
unfortunately, I can’t answer any of your other
questions. 

;)take care! 
lisaDAWNcolvin




__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com