[DL] Quantification over roles

[Franz Baader]

This is not true! In terms of first-order logic, roles are binary predicates.

Quantification over roles would be quantification over binary
predicates, which means that you would be in higher-order logic. The
same is, by the way, true if you quantify over concepts.

On Wed, Mar 2, 2011 at 10:24 AM, Cristiano Longo <longo at dmi.unict.it> wrote:
> Hi, I'm a novice about roles, so take care about what I say. In general
> roles may be considered as univerally quantified formulae, for example "A(x)
> <- B(x); C(x)" corresponds to the first order formula "(forall x)(B(x) and
> C(x) -> A(x)".
>
> Reasoning about roles themselves sounds like higher-level reasoning, which
> is not strictly related to quantifiers. May you provide some example to
> clarify?
>
> Cristiano Longo
>
>
>
> Il 01/03/2011 06:20, Steve W ha scritto:
>
> Hi,
> Is there any variant of DL out there that allows quantification over roles?
> I believe that's quite different to restriction of a concept by a role. If I
> want to reason over roles themselves, e.g., whether there exists a
> particular role such that some property holds, I'd need quantification over
> roles -- is that right?
> Thanks in advance for any input.
> Regards,
> Steve
>
> ---
> **  You received this mail via the description logic mailing list; for more
> **
> **  information, visit the description logic homepage at http://dl.kr.org/.
> **
>
>
> ---
> **  You received this mail via the description logic mailing list; for more
>  **
> **  information, visit the description logic homepage at http://dl.kr.org/.
>  **
>
>



-- 
--------------------------------
Prof. Dr.-Ing. Franz Baader
Technische Universität Dresden
Fakultät Informatik
Institut für Theoretische Informatik
Lehrstuhl für Automatentheorie
01062 Dresden
Tel.: +49 (351) 463-39160
Fax: +49 (351) 463-37959
E-Mail: baader at tcs.inf.tu-dresden.de
--------------------------------