ONTONYM Ontologies for pervasive computing

Modelling Events

The event ontology provides a means of describing activities of interest within a system. This could be something as simple as a door opening, or a complex notion or a conference made up of multiple sub events, with participants playing roles.

We define an event broadly as something which has a temporal dimension. The InstantEvent and IntervalEvent classes can be used to model events (see Feng Pan and Jerry R. Hobbs. Time in OWL-S. In Proceedings of the AAAI Spring Symposium on Semantic Web Services, Stanford University, CA, pp. 29-36. 2004.

The temporal predicates atTime and timeSpan to specify event boundaries. A further two classes SpatioInstantEvent and SpatioIntervalEvent represent temporal events with an associated location. The atLocation property from the Location ontology is used to link the two.

The event ontology also defines a Role class and containsRole property to define roles that are played by entities within the activity. A generic playsRole property then allows an entity (person, device, etc.) to be associated with an instance of an event. These concepts are designed for extension to describe domain-specific events. For example, we might define an ontology to describe presentations that contains

• a subclass of Event called Presentation
• two subclasses of Role: Speaker and AudienceMember
• sub-properties of the playsRole property for each role class (speaker, attendee)

We could then define an instance of a presentation as follows:

  :bobsPresentation a presentation:Presentation .
    rdfs:comment "Bob's presentation about Fossils to CSI" @en ;
    event:timeSpan
      [ a time:Interval ;
        time:hasBeginning "2009-05-12T13:00:00Z"^^xsd:dateTime ;
        time:hasEnd "2009-05-12T14:00:00Z"^^xsd:dateTime ] .
 :Bob presentation:speaker :bobsPresentation ;
 :Alice presentation:attendee :bobsPresentation ;
 :David presentation:attendee :bobsPresentation ;
 :Paula presentation:attendee :bobsPresentation ;

If implementing the model in software, we can use rules to infer that Bob has the type of Speaker and that attendees have the type AudienceMember during the period covered the event.