Start Date
11-12-2016 12:00 AM
Description
Diagrammatic conceptual schemas are an important part of information systems analysis and design. For effectively communicating domain semantics, modeling grammars have been proposed to create highly expressive conceptual schemas. One such grammar is the Web Ontology Language (OWL), which relies upon description logics (DL) as a knowledge representation mechanism. While an OWL DL diagram can be useful for representing domain semantics in great detail, the formal semantics of OWL DL places a burden on diagram users. This research investigates how user’s prior knowledge of the application domain impacts solving inference tasks as well as schema-based problem-solving tasks using OWL DL diagrams. Our empirical validation shows that application domain knowledge has no effect on inference performance but enhances schema-based problem-solving performance. We contribute to the conceptual modeling literature by studying task performance for a highly expressive modeling grammar and introducing inference tasks as a new task type.
Recommended Citation
Leukel, Joerg; Sugumaran, Vijayan; and Hubl, Marvin, "The Role of Application Domain Knowledge in Using OWL DL Diagrams: A Study of Inference and Problem-Solving Tasks" (2016). ICIS 2016 Proceedings. 8.
https://aisel.aisnet.org/icis2016/GeneralIS/Presentations/8
The Role of Application Domain Knowledge in Using OWL DL Diagrams: A Study of Inference and Problem-Solving Tasks
Diagrammatic conceptual schemas are an important part of information systems analysis and design. For effectively communicating domain semantics, modeling grammars have been proposed to create highly expressive conceptual schemas. One such grammar is the Web Ontology Language (OWL), which relies upon description logics (DL) as a knowledge representation mechanism. While an OWL DL diagram can be useful for representing domain semantics in great detail, the formal semantics of OWL DL places a burden on diagram users. This research investigates how user’s prior knowledge of the application domain impacts solving inference tasks as well as schema-based problem-solving tasks using OWL DL diagrams. Our empirical validation shows that application domain knowledge has no effect on inference performance but enhances schema-based problem-solving performance. We contribute to the conceptual modeling literature by studying task performance for a highly expressive modeling grammar and introducing inference tasks as a new task type.