ModelCorrect: Automated Correction of Visual Models during Computer Science Education
|
Funding: BMBF Period: 2019 - 2020 Greater project: bologna.lab The successful transfer of educational contents needs the active involvement of students. In computer science education at universities, exercises usually serve to deepen and apply knowledge taught in the lectures. Due to a lack of human resources, exercises often are held without a correction of exercise sheets. Hence, the students miss valuable feedback. For quite some time, computer-aided exercise systems are in use to compensate for this loss. Available systems which are tailored for computer science are suitable for the automated evaluation of coding tasks, but they insufficiently support software and systems modeling which is of high priority in computer science education. The automated correction of exercises requires the comparison of student solutions with a given standard solution. For this purpose, generally applicable methods for model comparison have become available recently. The project aims to adapt model comparison algorithms for a range of important model types, to integrate them with suitable visualization techniques and to provide them within a computer-aided exercise framework. This enables individual feedback and thus an expanded training of modeling competences which are considered as one of the key skills especially in the field of software engineering. |
