W3C home > Mailing lists > Public > public-hcls-coi@w3.org > July to September 2008

RE: Multi-layered Knowledge Representations for Healthcare (was RE: An argument for bridging information models and ontologies at the syntactic level)

From: Olasov, Ben <olasov@medicine.ucsf.edu>
Date: Thu, 24 Jul 2008 15:53:27 -0700
Message-ID: <16779E5DA610CF4BBCE391F0A64C78CF05AC3954@EXVS06.net.ucsf.edu>
To: dan.russler@oracle.com, "Samson Tu" <swt@stanford.edu>
cc: "Kashyap, Vipul" <VKASHYAP1@PARTNERS.ORG>, "Elkin, Peter L., M.D." <Elkin.Peter@mayo.edu>, public-semweb-lifesci@w3.org, public-hcls-coi@w3.org

> From: public-semweb-lifesci-request@w3.org on behalf of Dan Russler
>  
> Hi Samson,
>
> If "denote" = "describe" in your sentence, then I withdraw my objection.

But "denote" /= "describe".  These two words were intended to be, and are, used differently.  Consider the sentence, "The matrix P contains the entries pij, denoting the probability that for a speaker object i is associated with sound j.".  The word "denoting" functions here in the sense of "marking" or "signifying".  The word "describing" has no equivalent sense that would enable it to be substituted for "denoting" in this sentence.

I think we need to be very careful about assumptions of denotational equivalence, especially when the words under consideration are the very ones that make the idea of denotational equivalence possible and useful in the first place.

> My concern is that the term "class" as used in UML doesn't seem to mean
> the same thing as you are describing for a class in OWL. For instance, I
> don't see the same concept of "resource" in the definition of class in UML.
>
> A UML static diagram is just a symbolic method of displaying a set of
> related assertions, i.e. attribute assertions, association assertions,
> and state transition (behavior) assertions. The semantic interpretation
> of what the class means comes not only from the text definition of the
> "class symbol," but also from inferences made from the entire network of
> attributes, associations, and state transitions.
>
> To infer from a UML class more than is implied by the the above
> statement is incorrect.
>
> If we can agree on that, then we can better evaluate the equivalent and
> non-equivalent semantics of OWL and the other methods for organizing
> sets of assertions.
>
> Dan
>
> Samson Tu wrote:
>
> >
> >
> > Dan Russler wrote:
> >
> >> Hi Samson,
> >>
> >> We are getting closer.
> >>
> >> 1) In the reference you site..."A class is the descriptor for a set"...
> >>
> >> 2) Earlier, you stated that "semantics of a class as denoting a set
> >> of instances."
> >>
> >> I believe these two statements represent the "apples" and "oranges"
> >> you referenced:
> >>
> >> Statement 1) is the traditional "a class describes the attributes and
> >> associations for a concept that are common to a set of instances."
> >> Statement 2) is better described by your population example.
> >>
> >> I wasn't objecting to 1) . I was objecting to your seeming to
> >> confusie the 2) with 1).
> >>
> >> However, if you claim that "denote" means the same thing as
> >> "describe," then I would agree with you instead of objecting to your
> >> assertion.
> >>
> >> To be a little clearer....The definitions in a set of dictionaries
> >> all "describe" the meaning of the word "farmer." However, the word
> >> "farmer" in a dictionary does not "denote" the set of instances of
> >> farmers in the world. Same with a UML class titled "farmer."
> >
> > Dan,
> >
> > Yes, I am claiming that "denote" means the same thing as "describe" in
> > my intended usage of the English words.
> >
> > We are talking about the semantics of "class." The class "farmer" is
> > not the same thing as the dictionary word "farmer." Some people say
> > that UML is just a graphical notation without semantics because it
> > does try to make its meaning of the word "class" very clear. In
> > logic-based knowledge representation languages, the set-theoretic
> > semantics of class is widely used.
> >
> > The OWL Reference[1] put it this way:
> >
> > Classes provide an abstraction mechanism for grouping resources with
> > similar characteristics. Like RDF classes, every OWL class is
> > associated with a set of individuals, called the class extension. The
> > individuals in the class extension are called the instances of the class.
> >
> > OKBC [2] p. 6 put it even more baldly:
> >
> > A class is a set of entities. Each of the entities in a class is said
> > to be an instance of the class.
> >
> > In logical term, a class is a unary predicate satisfied by all of its
> > instances. Dictionary definitions of words are not involved.
> >
> > [1] http://www.w3.org/TR/owl-ref/#Class
> > [2] http://www.ai.sri.com/~okbc/okbc-2-0-3.pdf
> >
> >--
> >---------
> >Samson Tu                                   email: swt@stanford.edu
> >Senior Research Scientist                   web: www.stanford.edu/~swt/
> >Center for Biomedical Informatics Research  phone: 1-650-725-3391
> >Stanford University                         fax: 1-650-725-7944

Ben
Received on Friday, 25 July 2008 00:13:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 25 July 2008 00:13:29 GMT