Prof. Dr. Christoph Schmitz: In computer science courses, exercises and exams are often created that deal with the modeling of data and processes. Examples include UML and entity-relationship models, queries in SQL or data flows in big data systems. Models are usually derived from natural language descriptions of specific scenarios. In this initiative, software is developed to automatically generate variants of such tasks using artificial intelligence methods.
01 FEBRUARY 2022
It is based on a model library that is specified. The elements of the model can be added to subsets, which make it possible to group model elements that are technically related.
The variants of the original model can be generated by excluding certain subsets or showing only certain subsets. Moreover, it is possible to formulate logical conditions on the model elements, thus capturing dependencies between them. This allows, for example, the realization that certain subsets only occur in combination or mutually exclude each other.
For each variant created this way, a textual description must then be produced representing the scenario of the corresponding model. These textual descriptions can then serve as assignments for students, with the corresponding model variants serving as their solutions.
You are leaving the official website of Trier University of Applied Sciences
