Ontology engineering is a difficult activity and requires both domain knowledge and expertise in modelling notations such as OWL. The symbolic-like nature of OWL and, even more so, description logics can pose a barrier to entry for those who are not mathematically trained. By contrast, diagrams are often seen as user-friendly notations that can be accessible to broad user bases. Indeed, recent research has shown that diagrams support cognitive offloading as compared to symbolic and sentential notations. The accessibility of diagrams motivated the development of concept diagrams which were specifically designed for ontology engineering. In this talk I will introduce concept diagrams which have a formal syntax and semantics. I will then present recent research which has empirically identified diagrammatic patterns that allow people to perform tasks most accurately or quickly. These patterns are for commonly occurring, simple constructions such as AllValuesFrom axioms. I will then present results of a second empirical study which compared these diagrammatic patterns to both OWL and description logic. The results found, for the tasks undertaken, that participants performed significantly more accurately and significantly faster with the diagrammatic patterns that when using either OWL or description logic. Moreover, OWL significantly outperformed description logic. I will end the talk with a brief look at some research challenges that must be addressed to make concept diagrams a viable option for ontology engineers to use in practice.   This research is in collaboration with Eisa Alharbi, John Howse and Ali Hamie.

Watch the webcast replay >>