Hi everyone -

I’ve recently run into a situation where I’m not clear on whether my
problem is a limitation of the tool (or my misuse of it) or one of my
understanding in the modeling. Here’s the gist.

I want to extend rdf:Property in order to add associations for properties
like data classification (PII, HBI, etc.). Hence, I created the following:

:Property a rdfs:Class ;
        rdfs:label "Property" ;
        rdfs:comment "A managed value in the graph." ;
        rdfs:subClassOf rdf:Property .

:classification a rdf:Property ;
        rdfs:label "classification" ;
        rdfs:comment "Data classification for a property." ;
        rdfs:range xsd:string;
        rdfs:domain :Property> .

My thinking is that based on the default reasoners in Jena, the RDF
inferencing rule for rdfs:subClassOf should yield 2 rdf:type triples (one
explicit, one inferred) for the following:

:value a :Property;
        rdfs:label "value";
        rdfs:comment "The value of the quantitative value or property value
node.";
        rdfs:range xsd:string, xsd:decimal;
        rdfs:domain :MonetaryAmount .

Again, my expectation is that I should have 2 triples:

:value rdf:type :Property
:value rdf:type rdf:Property

However, I never see the inferred triple in arq.

Am I missing a configuration step for arq, thinking about my model
incorrectly, or something else entirely?

Thanks,

_howard

Reply via email to