Open Science
Our research group works predominantly with open-source software and Linux as the primary operating system. This is not a dogmatic choice but a practical one: open tools tend to be more transparent, reproducible, and better suited for research that others can build on.
We are happy to release code, data, and models from our research whenever the legal framework and cooperation agreements allow it. Our open-source releases typically use permissive licenses such as BSD, MIT, or Apache 2.0, as these tend to simplify adoption by industry partners. Where possible, we also publish open data to support reproducibility and enable others to build on our work. That said, we have great respect for copyleft approaches — the right license always depends on the context and goals of a project.
For book-related source code and errata, see Books.
Datasets
SELAQ — Student Expectations of Learning Analytics (2024)
Anonymized responses to the Student Expectation of Learning Analytics Questionnaire (SELAQ) in 2022 and 2023. Published at LAK 2024, Kyoto.
Goal-shot — Benchmark for Ill-Posed Problems (2019)
Dataset accompanying the paper "Goal-shot, a Benchmark Problem for Ill-Posed Problems".
Ill-Posed Optimisation — Example Data (2019)
Supplementary data for "A Learning Approach for Ill-Posed Optimisation Problems" (AusDM 2019, Best Paper Award).
Webis-SDMbridge-12 — Simulation Data Mining Bridge Models (2012)
Corpus of bridge simulation data for data mining research. Published at AusDM 2011, Ballarat.
Code
Code repositories will be listed here as they become available.