Re: Modeling a specific construct - please help

On 13 Sep 2009, at 21:56, Pat Hayes wrote:

>
> On Sep 11, 2009, at 5:55 AM, Uli Sattler wrote:
>
>> this is a difficult one:
>>
>> - I assume that Ab, B, and C are individuals, and that 'preceeds'  
>> is 'directly preceeds' (otherwise, you should *not* conclude that A  
>> is ConcurrentWith C.
>>
>> - you can introduce a transitive superproperty 'preceeds-trans' of  
>> preceeds and find all instances of the class (e.g., via OWL API and  
>> reasoner or via the DL query tab in Protege 4):
>>
>> (preceeds-trans value C) or (Inv(preceeds-trans) value C)
>>
>> if A is *not* in the answer to this query, then you can assume that  
>> it is ConcurrentWith C.
>
> Well, no, you cannot (validly) conclude this. This is a non- 
> monotonic inference, which is not supported by the OWL semantics.  
> While it may work in particular cases where you know that your data  
> is complete in the required sense, it is not good practice to use  
> such inference patterns in OWL, as they will (not may, but WILL)  
> break in some cases. Think building a glass building over a known  
> seismic fault.
>
> Pat Hayes

Well really, what I would conclude, in any case, is that "A _may_ be  
concurrent with B" -
i.e. the OWL model does not rule it out.
It seems dangerous to conclude that "A is concurrent with B".  And it  
only holds within whatever scope
the OWL model (I deliberately do not say "ontology") holds.

The difficulty with such reasoning patterns is that they only work  
when you can
complete the knowledge base so that it is fully constrained.
In most of our biomedical models, we can rarely be certain enough
that all possibilities have been covered to reason that the only
possibilities left over are true, only that they might be and may
]be worth further investigation.

Although we have sometimes used this kind of reasoning on very  
restricted
data entry problems with multiple constraints where we can be sure  
that they
can all be covered.  In those cases the non-monotonicity is an  
advantage,
although I would try to confine it to increasingly large queries  
rather than
the KB itself. As we learn more, we add it to the query, so that query  
gets larger and the number of
possible answers to the remaining questions gets fewer.

This sort of reasoning used to be supported in the Protege Query tab,  
but is no longer.

Regards

Alan

>
>>
>> Cheers, Uli
>>
>>
>> On 11 Sep 2009, at 12:27, Milan Zdravkovic wrote:
>>
>>> I am a beginner in OWL, working on specific process ontology. I  
>>> have a problem in inferring the concurrency of process activities,  
>>> for example - on basis of asserted A preeceds B and C preceeds B,  
>>> I need to infer that: A isConcurrentWith C.
>>> I was trying with modeling domain of isConcurrentWith with  
>>> (Manchester syntax):
>>> preceeds some (suceeds min 2 ProcessActivity)
>>> , where preceeds properties are asserted and suceeds is inferred  
>>> inverse property, but without success.
>>>
>>> Could you please help me on this ?
>>>
>>>
>>>
>>
>>
>>
>
> ------------------------------------------------------------
> IHMC                                     (850)434 8903 or (650)494  
> 3973
> 40 South Alcaniz St.           (850)202 4416   office
> Pensacola                            (850)202 4440   fax
> FL 32502                              (850)291 0667   mobile
> phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
>
>
>
>
>
>

-----------------------
Alan Rector
Professor of Medical Informatics
School of Computer Science
University of Manchester
Manchester M13 9PL, UK
TEL +44 (0) 161 275 6149/6188
FAX +44 (0) 161 275 6204
www.cs.man.ac.uk/mig
www.co-ode.org
http://clahrc-gm.nihr.ac.uk/