Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-owl] A question on Aggregation characteristics of SWRL

Thomas Russ tar at ISI.EDU
Thu Oct 30 12:22:55 PDT 2008


On Oct 30, 2008, at 9:15 AM, Matsokis Aristeidis wrote:

> Dear all,
>
> I have the SWRL rule:
> Corrective_Maintenance(?x) ^hasValue(?x, ?v)  -> sqwrl:sum(?v)
>
> On the SWRL Tab the result is shown.
>
> However, it seems like there is no way I can transfer this  
> information in an OWL instance directly, meaning that I cannot do  
> something like this:
> Corrective_Maintenance(?x)   
> ^Results(Results_Corrective_Maintenance)  ^hasValue(?x, ?v)  ->  
> sqwrl:sum(?v) ^ hasSum (Results_Corrective_Maintenance, sqwrl:sum(?v))

That is correct.  You cannot do this.
>
> Have I missed something here?
> Is there any alternative way (other than manually or through csv- 
> >excel-> DataMaster) I could transfer this information to an OWL  
> instance (Results_Corrective_Maintenance)?

You would have to do it manually or through program code.

That because the aggregation operators make an implicit closed-world  
assumption and thus cannot be used for OWL inferences, since the OWL  
semantic model specifies open-world semantics.  Allowing such closure  
would violate the semantic model, and it is therefore not supported by  
inference tools such as SWRL, which want to preserve the OWL semantics.

> If none of the above is valid, is there a plan to implement such  
> support?
>
> Thank you very much in advance for your time once more.



More information about the protege-owl mailing list