Skip to content

About

Project

This package was developed as part of a master’s thesis in Cognitive Science at Aarhus University. It provides a modular framework for systematically capturing and analyzing interactions between students and LLM-based chatbots in educational settings.

The core contribution is supporting quantitative, interaction-level evaluation by turning dialogue data into structured, labeled outputs. By combining tools for classification, simulation, and visualization, the package enables data-driven reproducible analysis of chatbot-assisted learning and is intended for use in both experimental and applied research contexts.

The full source code for the EduChatEval package is available on GitHub and is licensed under MIT License.

Contact

For questions or collaboration inquiries, feel free to reach out to Laura W Paaby.

Acknowledgements

This project builds on existing tools and ideas from the open-source community. While specific references are provided within the relevant scripts throughout the repository, the key sources of inspiration are also acknowledged here to highlight the contributions that have shaped the development of this package.