Re: Support for string facets in DL Query/Property Restrictions

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Re: Support for string facets in DL Query/Property Restrictions

Timothy Redmond

Thanks Nick for the correction.  I was not aware of the OWL 1.1  
support for datatypes.  Oops...

-Timothy

On May 2, 2008, at 12:44 AM, Nick Drummond wrote:

> Support in the OWL API and therefore P4 has been following reasoners
> by exposing this functionality as it becomes available.
> The datatype support matrix [1] currently suggests (unless I'm reading
> it wrong) that these facets are not yet implemented in our current
> reasoners.
>
> Of course, the expression parser (using the OWL API parser) should
> support these valid facets anyway (but you will not get any benefit in
> querying the reasoner until it is supported.
>
> Nick
>
> [1] http://spreadsheets.google.com/pub?key=pTmcCXR-dV6RpTEPxB0O-DQ
>
> On 1 May 2008, at 17:53, Timothy Redmond wrote:
>
>>
>> I don't think that this is going to be implemented soon.  Part of the
>> issue  is the difference between search and inference.  Searching can
>> be settled just be sifting through the syntactic representation of  
>> the
>> ontology.  But searching may not  see all the facts that can be
>> deduced.
>>
>> The most obvious way to try to implement your functionality would be
>> to have reasoners  understand  the string datatype much better.  So
>> for instance to determine that a query like
>>
>>   (p all string[maxlength 2]) and (p all string[contains, "Pizza"])
>> and (p some [string])
>>
>> is unsatisfiable, the inference  engine would have to deduce that any
>> string containing "Pizza" must have more than two characters.  I  
>> don't
>> know if anybody is thinking about this work.  (Actually it would seem
>> that it ought to be possible to make tableaux reasoners extensible to
>> handle any datatype that has a complete decision procedure of its
>> own.  But I am not sure if this is true.)
>>
>> -Timothy
>>
>>
>> On Apr 28, 2008, at 5:26 AM, Liu, Richard wrote:
>>> Hello,
>>>
>>> In my ontology I have a class Report and a data property hasTitle
>>> with domain Report and range string (xsd:string).  In the DL Query
>>> tab I would like to find all the individuals in Report that have a
>>> title with a given regexp pattern.  I think this should be possible
>>> with a query like “hasTitle some string[pattern ‘regexp’]”, but the
>>> DL Query parser seems not to “like” any facet of “string”.  I also
>>> tried “minLength 2” instead, but that had the same problem.  In both
>>> instances DL Query seems to expect a Property after the string.
>>> Hoping to circumvent the problem, I tried to define a subclass of
>>> Report with the necessary and sufficient condition “hasTitle some
>>> string[pattern ‘regexp’]”, but that proved unacceptable as well.  It
>>> doesn’t even seem to work in Protégé 3.4 beta.  Either I am
>>> misunderstanding something that is very fundamental – not an
>>> unreasonable assumption – or these facets are not (yet?)  
>>> implemented.
>>>
>>> Thanks,
>>> Richard
>>> ____________________________
>>>
>>> Richard Liu
>>> Group Research Information
>>> Scientific Information Resources
>>> F. Hoffmann-La Roche Ltd.
>>>
>>> PXIBR
>>> Bldg. 092/6.01
>>> CH-4070 Basel, Switzerland
>>> Phone: +41 (0)61 687 92 14
>>> FAX: +41 (0)61 688 95 70
>>> [hidden email]
>>> Confidentiality Note: This message is intended only for the use of
>>> the named recipient(s) and may contain confidential and/or
>>> privileged information. If you are not the intended recipient,
>>> please contact the sender and delete this message. Any unauthorized
>>> use of the information contained in this message is prohibited.
>>>
>>>
>>> _______________________________________________
>>> p4-feedback mailing list
>>> [hidden email]
>>> https://mailman.stanford.edu/mailman/listinfo/p4-feedback
>>
>> _______________________________________________
>> p4-feedback mailing list
>> [hidden email]
>> https://mailman.stanford.edu/mailman/listinfo/p4-feedback
>
> _______________________________________________
> p4-feedback mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/p4-feedback

_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback