Re: [protege-user] Some doubts

>
> I am trying to build an ontology with a class, "Letters," containing two
> subclasses: "Vowels" and "Consonants." I create the enumerated class
> "Vowels" as in exercise 24 of the manual dated October 8, 2021, by adding
> the instances A, E, I, O, U and then accessing "Equivalent to".


What are you making equivalent to what... and why?  I don't understand
based on your description why you need to use Equivalent To and what you
are making Equivalent To what. IMO people shouldn't use Equivalent To (or
SameIndividual) for most use cases. If you have a class or individual that
has more than one name, using skos:altLabel and skos:prefLabel are usually
much better options. Less overhead and you don't make the reasoner do extra
work for little gain. Typically, I only use Equivalent To when I'm
combining two ontologies. E.g., in Dublin Core and Prov there are two Agent
classes and for a use case I'm working on, I wanted to use both but I only
wanted one Agent class so I made them Equivalent To each other.

Or for a training ontology I created two instances: one called Jay_Gatsby
and called James_Gatz -- in Fitzgerald's novel The Great Gatsby, Gatz was
Gatsby's real name but he was ashamed of his humble roots so people in the
novel knew him as either Gatz or Gatsby but except at the end no one knew
Gatz and Gatsby were the same Person. I thought that was a good example of
a scenario where SameIndividual made sense.

And another question: How could I use Protege to model the basic questions
> of set theory?


What basic questions are you referring to? If you mean can you model things
like Russell's Paradox then the answer is no. OWL is an implementation of
Description Logic not First Order Logic so (by design) issues that apply to
FOL (e.g., the Halting aka Entscheidungs Problem: no language with the
power of FOL can have a reasoner guaranteed to terminate as proven by
Turing and Church) can be modeled. But if you mean can you model things
like intersection, union, etc. then absolutely yes. The semantics of OWL
are the semantics of set theory and Description Logic and Description Logic
is the result of a fair amount of knowledge representation research where
people wanted to find a language close to the expressive power as FOL that
is still decidable.

If you send your ontology I can take a look and probably figure out what's
going on pretty quickly. BTW, not sure what tutorial you are using but you
might find my revision of the Pizza tutorial worth doing. Besides building
the actual Pizza ontology I also added chapters for SWRL, SPARQL SHACL and
Web Protege: https://www.michaeldebellis.com/post/new-protege-pizza-tutorial

Cheers,
Michael
https://www.michaeldebellis.com/blog

On Wed, Nov 12, 2025 at 3:10 AM Joaquín José Castelló Benavent via
protege-user <[email protected]> wrote:

> I am trying to build an ontology with a class, "Letters," containing two
> subclasses: "Vowels" and "Consonants." I create the enumerated class
> "Vowels" as in exercise 24 of the manual dated October 8, 2021, by adding
> the instances A, E, I, O, U and then accessing "Equivalent to".
>
> Next, I try to do the same with the "Consonants" class, but when I go to
> add the instances, the vowels I added earlier appear. If I remove them and
> add the consonants (B, C, D, ..., X, Y, Z) and finish the process with
> "Equivalent to," the vowels disappear when I synchronize, and I can't get
> both classes at the same time.
>
> What am I doing wrong?
>
> And another question: How could I use Protege to model the basic questions
> of set theory?
>
> Thank you very much in advance, and please excuse any inconvenience this
> may have caused.
>
> Yours sincerely
> Ximo Castelló.
>
> _______________________________________________
> protege-user mailing list
> [email protected]
> https://mailman.stanford.edu/mailman/listinfo/protege-user
>

_______________________________________________
protege-user mailing list
[email protected]
https://mailman.stanford.edu/mailman/listinfo/protege-user