Automated Assessment of Conceptual Models in Education A Systematic Literature Review

In Computer Science, Software Engineering, Business Informatics or Information Systems, conceptual modeling is an important tool and as such also contained in the respective curricular recommendations. Especially in large university courses, an automated assessment of models can improve the quality of teaching and learning. While there are many different approaches to automatically assess conceptual models, these approaches, however, often only tackle a single aspect or a single type of conceptual model. In this paper, we aim to take a comprehensive perspective on the topic and shed light on the current state of the art and technique. Furthermore, as assessment approaches have to be developed in accordance with appropriate teaching or learning activities and desired learning outcomes, we inquire in which settings automated assessment approaches are included and to which extent didactic aspects are taken into account. To this end, we have conducted a systematic literature review in which we identified 110 relevant publications on the topic which we have analyzed in a structured way. The results provide answers to five relevant research questions and pinpoint open issues which should be inquired in further research.