W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2015

Re: Conservative Extension of OWL 2

From: Ignazio Palmisano <ipalmisano.mailings@gmail.com>
Date: Tue, 11 Aug 2015 20:13:46 +0100
Message-ID: <CAD2jOMNrHkpO8+5ERPvfrAYzokh-JBRAeVnj2Ou-2ACa0MW7LQ@mail.gmail.com>
To: "public-owl-dev@w3.org" <public-owl-dev@w3.org>
On 11 August 2015 at 17:44, Leila Bayoudhi <bayoudhileila@yahoo.fr> wrote:
> HI?
> Thanks for your answer.
> I noticed that this notion a a strong relation with "ontology
> modularization". Has it a relation with ontology evolution?

In a way, yes, as extending an ontology is a form of evolution.
However a conservative extension is a very specific form of evolution.
I don't know if anyone studying ontology evolution has connected it
formally to conservative extensions.

> Is it expected to have a conservative extension of the ontology since it
> relies on logic while evolving an ontology?

Conservative extensions are rare and hard to build, so I'd say no -
not necessarily. Plus, ontology evolution, on occasion, /wants/ to
change the meaning of an existing ontology. Conservative extensions
never do that. So, in that sense, they're incompatible.


> Le Lundi 10 août 2015 22h53, Ignazio Palmisano
> <ipalmisano.mailings@gmail.com> a écrit :
> On 10 August 2015 at 22:21, Leila Bayoudhi <bayoudhileila@yahoo.fr> wrote:
>> Hi,
>> I have some questions:
>> -What is exactly a " consrvative extension of an ontology"?
> In short, A is a conservative extension of B if all entailments of B
> are also entailments for A. However, the concept is much more complex
> than this. You can start with
> http://cgi.csc.liv.ac.uk/~frank/publ/ijcai07.pdf, and move on to the
> large body of literature on the topic.
> Approximations of conservative extensions are also an important
> research area. Many modularisation techniques rely on them. See for
> example http://www.cs.man.ac.uk/~sattler/publications/chks07-just.pdf
> (These are pointers to some of the oldest literature on the topic.)
>> -Does ontology is expected to have conservative extension since it relies
>> on
>> logic?
> I'm not clear on what you mean here.
>> -Does OWL 2 enables this feature?
> No. Building conservative extensions is not enabled by OWL 2. It is a
> complex task (undecidable for high expressivity ontologies).
> Approximations, like modules, can be built without OWL 2 (they predate it).
> HTH,
> I.
>> thx for answering me.
Received on Tuesday, 11 August 2015 19:14:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 11 August 2015 19:14:15 UTC