392218 Developing a task editor feature in an Intelligent Tutoring System (Pj) (WiSe 2024/2025)

To support programming education, the research group "knowledge representation and machine learning" has developed an Intelligent Tutoring System. In this project, we invite students to contribute to this ITS by adding new features for teachers. More precisely, we want teachers to be able to fix, edit and create new programming tasks and courses directly within our system. In its current stage, courses are created offline using a specified format and are parsed via a script. In the course of this project, we would like to streamline this process and develop prototypes for (AI-based) support-tools for course and task development.

More specifically, potential tasks are:
- Creating a user-interface for navigating and editing programming tasks online.
- Allowing testing the tasks online.
- Automated unit test generation for the new tasks using, for example, large language models.
- Providing automated, pedagogical feedback on task descriptions using, for example, large language models.
- Own ideas are welcome!

The implementation of these tasks intends to help fulfill following educational goals:
- Learning to deploy and improve existing systems in development.
- Get a hands-on experience with Angular and Python-based web applications.
- Learning how to understand the needs of target users (i.e. teachers) and translate these needs into technical tasks/goals.
- Learning to work on an ongoing software development project with several independent collaborators.
- Contributing to ITS research.
- Planning and executing an evaluation study.

What we offer:
- Participation in a real-world project that will be implemented in university courses and research.
- Advice in bi-weekly development meetings.
- Technical support for the deployment and on-going work.
- Creative freedom for designing the features.

Bibliography

Literature:
- Alina Deriyeva, Jesper Dannath, Benjamin Paaßen (2024).
Relation between struggle and learning personality in programming exercises. 22. Fachtagung Bildungstechnologien (DELFI), Fulda, Germany, in press.
- Jesper Dannath, Alina Deriyeva, Benjamin Paaßen (2024).
Evaluating Task-Level Struggle Detection Methods in Intelligent Tutoring Systems for Programming. 22. Fachtagung Bildungstechnologien (DELFI), Fulda, Germany, in press.