Teaching Chemometrics and ML for Analytical Chemistry

We develop and evaluate innovative teaching approaches that lower the entry barrier to coding while strengthening chemometric and machine learning (ML) competence in Analytical Chemistry. Our strategy combines Orange (a visual programming environment for ML) with Jupyter/Notebook-style materials that promote reproducibility, literate computation, and good scientific programming habits.

Chemometrics and ML are increasingly essential in modern Analytical Chemistry, yet they remain challenging to teach because students often face a steep learning curve: programming anxiety, fragmented toolchains, and difficulty linking abstract algorithms to real analytical decisions. We address this by building hands-on, scaffolded learning modules where learners move seamlessly from graphical workflows to editable code, using authentic datasets (e.g., spectroscopy and chromatography) and emphasizing core analytical thinking: preprocessing choices, calibration/validation, model interpretation, and responsible reporting.

A distinctive element of this line is evidence-based educational design: we assess usability and learning outcomes with structured instruments (rubrics, surveys, and performance-based tasks), iteratively refining the materials to maximize engagement, conceptual understanding, and transfer to real laboratory and research scenarios.