Glossary · Semantic interoperability and information models
Ontology mapping
German: Ontologieabbildung
In semantic technology, ontology mapping is the formal specification of how concepts, properties and instances of one ontology relate to or are transformed into those of another, for example as equivalence statements, SKOS mapping relations or transformation rules, so that data can be translated or queried across ontologies.
- Information models
In one sentence
Ontology mapping formally specifies how concepts and data of one ontology translate into another, for example with rules or SKOS relations.
Example
A mapping rule converts the property ratedPower in kilowatts from the supplier's ontology into the property nominal power in watts in the plant's ontology.
How it applies
- Engineering: Mappings make heterogeneous data usable together, for example when combining supplier data, engineering data and maintenance data in a Knowledge graph.
- Integration: Be explicit about the mapping type. An exact match, a close match and a broader match (as in SKOS) have different consequences for automated reasoning; overstating equivalence propagates errors.
- Documentation: Version mappings together with the ontologies they connect, and document unit conversions and value transformations. Test mappings with real data.
Ontology mapping vs. data mapping
Data mapping connects fields of two data formats, such as columns in two databases. Ontology mapping connects concepts with defined meaning and can be used for reasoning. In practice, ontology mappings often drive the underlying data transformations.