Automated AI-based Feedback in Simulation-Based Learning Environments (Project 4)
Project description
Current machine learning approaches have the potential to objectively and reliably capture human behavior based on predetermined criteria. Student behavior (for example, in interaction with simulated patients during medical interviews or during a practice lecture) can be recognized in this way and thus used to formulate feedback regarding their competence development.?Building on extensive preliminary work of the experts involved in the project, a machine learning algorithm will be further developed that analyzes human behavior in terms of its convergence with desired behavioral criteria.?The results thus obtained will be used to present students with automated, AI-based feedback on the behavior they exhibit in digitally-enriched face-to-face settings in different subjects (communicative behavior, e.g., in doctor-patient or parent-advice meetings).?The added value compared to human feedback lies, among other things, in the greater objectivity and subsequent scalability of the approach.?The Faculty of Applied Computer Science and the Faculty of 伟德国际_伟德国际1946$娱乐app游戏icine are primarily responsible for this project. In the third year of the project, the developments will be transferred to teacher training (Faculty of Humanities and Social Sciences).
?
?
Publications
2024
- Hallmen, T., Deuser, F., Oswald, N., & André, E. (2024). Unimodal multi-task fusion for emotional mimicry prediciton.? PDF
- Schiller, D., Hallmen, T., Don, D. W., André, E., & Baur, T. (2024). DISCOVER: A Data-driven Interactive System for Comprehensive Observation, Visualization, and ExploRation of Human Behaviour. arXiv preprint arXiv:2407.13408.
- Hallmen, T., Deuser, F., Oswald, N., & André, E. (2024). Unimodal Multi-Task Fusion for Emotional Mimicry Intensity Prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (S. 4657-4665). IEEE Computer Society. https://doi.org/10.48550/arXiv.2403.11879
2023
- Hallmen, T., Mertes, S., Schiller, D., Lingenfelser, F., & André, E. (2023). Phoneme-Based Multi-task Assessment of Affective Vocal Bursts. In D. Conte, A. Fred, O. Gusikhin, & C. Sansone (Hrsg.), Deep Learning Theory and Applications (Vol. 1875, S. 209–222). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-39059-3_14
2022
- Bauermann, M., Schindler, A.-K., & Rotthoff, T. (2022). Studienprotokoll: Einfluss von Mimik in medizinischen ad hoc-Anvertrauensentscheidungen. In D. Stoevesandt (Hrsg.), GMA 2022 - Gemeinsame Jahrestagung der Gesellschaft für 伟德国际_伟德国际1946$娱乐app游戏izinische Ausbildung (GMA) und des Arbeitskreises zur Weiterentwicklung der Lehre in der Zahnmedizin (AKWLZ) (S. 37-38). Halle: German 伟德国际_伟德国际1946$娱乐app游戏ical Science GMS Publishing House.? https://doi.org/10.3205/22gma050
-
Hallmen, T., Mertes, S., Schiller, D., & André, E. (2022). An Efficient Multitask Learning Architecture for Affective Vocal Burst Analysis. https://arxiv.org/html/2210.15754/.
Project coordinator
- Phone: +49 821 598 - 72050
- Email: thomas.rotthoff@med.uni-augsburgmed.uni-augsburg.de ()
- Room 06.005 (Building Faculty of 伟德国际_伟德国际1946$娱乐app游戏icine)
Project participants
- Moritz Bauermann
- Dr. Kathrin Gietl????????????
- Tobias Hallmen?????????????
- PD Dr. Dr. h.c. (UA) Karoline Hillesheim???????????
- Prof. Dr. Miriam Kunz??
- Sandra Littwin
- Pia Schneider?
- Prof. Dr. Thomas Rotthoff
- Prof. Dr. Elisabeth André
Participating chairs and institutions (伟德国际_伟德国际1946$娱乐app游戏 of Augsburg)

The project "Facilitating Competence Development through Digital Authentic and Feedback-Based Learning Scenarios" is funded within the framework of the funding announcement "Facilitating 伟德国际_伟德国际1946$娱乐app游戏 Teaching Through Digitization" (FBM2020) by Stiftung Innovation in der Hochschullehre.