How to use domain and range in OWL for ontological reasoning

Next episode: click here

‍

Hello, and welcome to another episode of the RDFox introductory series.

In this episode, we're going to be looking at inferring the domain and range for a particular relationship with an OWL axiom. In this case, we're going to be focusing on an axiom that discusses the ‘result_driver’ predicate.

So what do we mean by this? Well, whenever we have a triple that involves the ‘result_driver’ relationship, we can infer some information based on the existence of this relationship about the subject and object of this triple.

For example, we know that any subject of a triple that includes the ‘result_driver’ relationship must be an entity of the class ‘result’, and likewise the object of this same triple must be from the class ‘driver’.

Now we may not have this class information explicitly given in our data, so we can use this axiom to infer that information. Here we specify that the domain of this relationship is the class ‘result’, so anything in the subject position will be classified as a result and likewise the range anything in the object position, this will infer that entities given there belong to the class ‘driver’.

So we can see this depicted in the image below where we have a screenshot from the RDFox Web console where explicit data is shown in dark grey and inferred information is shown in blue.

At the top we have an example of a triple where we see ‘result_driver’ connecting an entity of ‘result’ to an entity that is a ‘driver’. From this and with our axiom we also see that the blue inferred information has added the triples ‘result_1 a result’ and ‘driver_hamilton a driver’. So just like that we have inferred the domain and range of this relationship. And of course, with this single axiom, this will apply to all of the instances where the ‘result_driver’ relationship is seen.

Now it's important to remember here, this is not about data validation. This is not going to be constraining your data. This is an OWL axiom, it will only infer new data. So if you want to validate your data or constrain what comes in any way, you'll have to look at SHACL or Datalog in the RDFox docs.

But if you'd like to learn more about reasoning and inference, check out our other videos where we go into much more detail. If you'd like to learn more about our axioms or see specific examples in detail, continue through the rest of our workshop material or sign up for our next workshop for free, available on our website.

‍