Geek asks and answers

OWL-DL: property limitations and / or domain range?

I have doubts about the correct use of the domain, range and limitations for determining the ontology. If I want to imagine that “the student should have only one identifier”, I think I could do one of the following:

  • Define the domain and range for the property (for example, :hasId rdfs:domain :Student; rdfs:range :Identification) and make it functional.

  • Define the property and create a constraint in the student class: "hasId exactly 1 identity."

In terms of semantics, do they mean the same thing? If so, is there a preferred option in terms of agreements or best practices? Finally, what about data type properties? Can I just define a domain / range and / or do I have to create class constraints to associate it with a property?

+3
source share
1 answer

OWL2 has quantitative restrictions, which means you can have form class expressions

& exist; pC

which denotes a class of individuals that are relevant for property p, which is of type C. Similarly, you can use the constraint

& FORALL; pD

, p ( ) D. , , .

& leq; n p.C
& geq; n p.C
= n p.C

, . ,

p C

, p - , C. ,

p D

, - p, D. , , . , p D

& ; & Sqsubseteq; & FORALL; p.D

, & top; ( owl:Thing, .. ) , p D. , . p C,

& ; & Sqsubseteq; & FORALL; 1.C

, :

, " ", , :

  • (: hasId rdfs: : ;: hasId rdfs: range: Identification) .

  • : "hasId 1 ".

, ? , ?

-, . 1 ,

x hasId y
x hasId z

, x , y z y , z ( ), . , OWL, , , .. . . , hasId , ,

& ; & Sqsubseteq; & leq; 1 hasId. & top;

- . , ,

sqsubseteq; = 1 hasId.Identification

, , , hasId Student Identification , , , hasId, , :

sqsubseteq; = 1 hasId. & Top;

, - , . , - (, ), (, ). - , :

    • {& sqsubseteq; = 1 hasId.StudentIdentification}
    • StudentIdentification

  • hasId {domain: Person, range: Identification}

( , , .)

+10

More articles:


All Articles