Hybrid Knowledge Bases: the Interplay between Domain-specific Models and Knowledge Graphs
The interplay between Knowledge Graphs and Domain-specific Modelling Languages was traditionally concerned with the challenge of “ontological commitment” - i.e., checking the ontological qualities of a language in order to enforce coherence and consistency in model contents. An alternative approach to coupling the two knowledge representation approaches will be promoted by this lecture, motivated by pragmatic requirements of externalizing and combining both human-readable and machine-readable representations, thus giving a novel interpretation to the seminal SECI knowledge conversion cycle. Agile modelling languages are treated here as schemas for knowledge that is amenable to linking, reasoning and publishing with the help of semantic technology - i.e., RDF graph databases, OWL axioms, SPARQL queries and endpoints. The presentation highlights benefits of combining the Agile Modelling Method Engineering Framework with the Resource Description Framework in order to build novel knowledge acquisition methods. The lecture will present results derived from applying and refining this idea in the ComVantage FP7 EU project and the EnterKnow PED Romanian research project, taking it to the point where it may inspire a novel “model-aware” software engineering method.
Lecture at NEMO2021
Date/Time: Thursday, July 22, 2021 at 12:15