LLM-based training using personalized case examples
Revolutionizing education: A case-based learning coach powered by LLMs transforms abstract concepts into interactive, personalized experiences!
Factsheet
- Schools involved School of Engineering and Computer Science
- Institute(s) Institute for Patient-centered Digital Health (PCDH)
- Research unit(s) PCDH / AI for Health
- Funding organisation Others
- Duration (planned) 01.11.2024 - 31.12.2025
- Head of project Prof. Dr. Kerstin Denecke
-
Project staff
Prof. Daniel Reichenpfader
Denis Sumin Moser -
Partner
BeLEARN
Universität Freiburg
Tobias Häberlein, Fernfachhochschule Schweiz
Situation
This project aims to revolutionize how students learn abstract concepts by leveraging Large Language Models (LLMs) to create personalized, interactive, and practical case-based learning experiences. By allowing students to specify their interests, goals, and learning needs, the system generates tailored scenarios that make complex ideas more relatable and applicable to real-world contexts. Initially focused on psychology students in methodology courses, the project explores how adaptive learning tools can boost engagement, comprehension, and outcomes. The findings will pave the way for innovative, scalable educational solutions that blend cutting-edge AI with modern pedagogical approaches.
Course of action
Students can let the Tutor know their personal areas of interest. Based on these preferences, the Tutor generates interactive case examples – such as the interpretation of confidence intervals in clinical trials – tailored to the learners’ professional goals. The Tutor guides the students through case-based simulations, provides corrective feedback and asks follow-up questions to deepen their understanding. The effectiveness of the Tutor is currently being evaluated in an experimental study at the University of Bern. This study compares commitment, skills development and satisfaction between two cohorts: those using the LLM Tutor versus those using traditional teaching methods. In parallel, the Tutor is being implemented in statistics courses at FernUni Schweiz, the FFHS and the University of Fribourg, where it is being evaluated through experimental or observational studies.
Result
By offering a wide variety of learning opportunities tailored to students’ interests, the Tutor helps students to understand and engage with abstract subject matter. Following successful trials at the University of Bern, the Tutor was deployed in three methodology courses: one at UniDistance Suisse, one at the FFHS and one at the University of Fribourg. It is available to the public as an open-source tool on GitHub.
