Project description
The dynamic digitalisation of our working world makes comprehensive digital skills indispensable. However, instead of teaching these in a targeted way, teaching at universities focuses more on the use of digital tools than on the actual understanding of digital topics. As a result, students are missing out on important skills they need to be successful in an increasingly digitalised world.
The interdisciplinary research project in cooperation between bidt and the Technical University of Munich addressed this gap. The project team explored the extent of digital learning content in higher education teaching by combining knowledge from the fields of computer science and educational science.
At the centre of the methodological approach was the analysis of course descriptions from current course catalogues or module handbooks from various universities. The project team used natural language processing (NLP) models for the analysis, which includes both already established NLP methods and the development of a special NLP method for the evaluation of course descriptions. This enabled an in-depth evaluation of the degree of digitalisation and the quality of the content conveyed.
The aim of the project was to develop a user-friendly website based on these findings that provides students with a comprehensive overview of digital teaching content at selected universities. In addition, the findings enabled a simple assessment of the extent and distribution of digitisation-relevant content in university courses, generated an overview of knowledge transfer and identified gaps in the teaching offer.
The research project successfully developed an NLP-based methodology for the automated analysis of module descriptions and quantified the degree of digitisation of teaching content at five German universities. Building on these findings, the system was further developed into a RAG-based recommendation system (beta version) that uses large language models to generate personalised module recommendations for students using natural language input [https://nerd.sse.cit.tum.de/].
The project was completed by 30 September, 2023. However, further development of the tool will continue at the Technical University of Munich until the end of 2026.
Contact
Research
Project team
Dr. Severin Kacianka
PostDoc Researcher, Chair of Software Engineering | Technical University of Munich
Prof. Dr. Alexander Pretschner
Chairman of bidt's Board of Directors and the Executive Commitee | Chair of Software & Systems Engineering, Technical University of Munich | Scientific director, fortiss
Alumni
