Hello,
The problem of the proper ontology occures, if you think, what kind of
users do you have, who query your dataset. If you think there are aliens,
then you have to define the whole world with your ontology whathever your
dataset represents.
If you make for me an ontology for dbpedia-
You can keep the Occupation tree under person if you can live with the
fact that these classes describe individuals, not occupations performed by
Persons. A Senator is a Politician, but a Parliament is not, because it is
an Organization. The Beatles is a Band but not an Artist, and the same is
I'll try to narrow the scope of the description of what I'm trying to build
and that I've published previously:
Besides some basic initial type inference I'm not focusing (only) on
ontology engineering or modeling but more in providing an engine (some like
DOM / JavaScript is for HTML) for whateve
Some thoughts.
(1) The foaf:Agent concept applies in many cases where people would want
to use a person. In my mind an "Agent" is something that can be the
originator or recipient of a communication thus it covers corporate
personhood, etc. That still leaves the question of what "The Dallas
Roles? (Actor / Role pattern in OOP). I call it 'metaclasses' using the
'kind' abstraction of type inference by aggregation of common properties
and their values. RDF allows for a class to be instance (not subclass) of
another class. Also OWL restrictions could be used to model roles.
Regards,
Seb
One reason to have a subclass structure under Person (but also under any
natural kind) is to support generalizations for subcategories. Many of the
subclasses of Person in DBpedia are for occupations, e.g., Architect, Athlete,
NascarDriver, Politician, and Senator. If these subclasses of Person a
