W3C home > Mailing lists > Public > semantic-web@w3.org > November 2007

Re: some problems of DL Reasoning

From: Bijan Parsia <bparsia@cs.man.ac.uk>
Date: Mon, 19 Nov 2007 23:49:34 +0000
Message-Id: <17EE5BF7-9292-487D-8D76-9006D0B4E216@cs.man.ac.uk>
Cc: "Jun Fang" <leon.essence@gmail.com>, semantic-web@w3.org
To: "Mauro Mazzieri" <mauro.mazzieri@gmail.com>

On Nov 19, 2007, at 2:14 PM, Mauro Mazzieri wrote:

>
> 2007/10/29, Jun Fang <leon.essence@gmail.com>:
>> I am new to the Description Logics. I have met the following problem.
>>
>> As we know, K |= C <==> C unsatisfiable with K,
>> here K is a DL knowledge base,

I.e., a set of *axioms*.

>> and C is a concept
>>
>> If K is set to {A,B},

That's not a set of axions.

>> and C is set to (AȁB),

I.e., ~A v B

>> then C is unsatisfiable with
>> K

I don't see why. The TBox is empty afaict. Thus ~A v B is satisfiable.

> C is incoherent with K: is a concept that can not have instances.

Why? What do you think K is?

> Toghere with ex. C(a) we have an inconsistent ontology.
>
>> it means K|=A subclassof B
>>
>> using the similar way, we  can also get K|=B subclassof A,

Doesn't work for me in the same was as above. Did something get munged?

>> I must be missing something obvious here. Can someone can tell me  
>> the reason
> As they're defined, A and B are the same class.
[snip]

You think "K is set to {A, B}" means "K = {A<->B}"? Ok. But, uh, if  
A<->B, then A->B and B->A....and neither is incoherent.

Jun, try downloading Protege4 or something similar and playing with  
real ontologies and a reasoner. I find it helps.

Cheers,
Bijan.
Received on Monday, 19 November 2007 23:54:39 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:47:29 UTC