3.7: Ontology Formalisation

Work in progress, not part of this version of the handbook

After the previous activities, we have made an inventory of CQ's, their concepts with definitions, considered a way to modularise and structure them, and identified potential ontologies for reuse. In order to be able to put the ontology to practical use, we now need to bring this all together and formalize it into a real ontology, i.e., define it in a language and formalism that allows unambiguous interpretation of the model by computer systems. This formal language provides a set of constructs that allow us to define our model. depending on the choice of language.

In order to formalize our model, we need to choose a formal language first. There are multiple options for formalization, each with their merits, advantages and limitations. Some examples include:

• RDF(S): a data modelling formalism, primarily developed and used in the semantic web community. It provides mechanisms for describing groups of related resources and the relationships between these resources.

• OWL:

The formalisation step also includes, besides adding necessary detail:

• First adding classes/objects and relations/properties. Add a definition, domain and range for each relation. Add a definition for each object.

• Keep in mind to reuse existing ontologies if possible. Or use as an inspiration.

• What tools to use and why? TopBraid? Protege?

In practice

• Note down all (technical) design choices in a structured manner. This prevents you from reinvestigating a design choice already made.

• Choose whether the IRI name and label name should be the same or should be different for some purpose.

• Add definitions and labels in different languages if necessary. This language extension can be part of the main ontology, but it can also be defined in a separate file. If people want to use this language extension, they can easily import the separate file.