A four-part module on machine learning in STEM education research using R


For the NSF-funded LASER Institute, I developed the initial materials last year on machine learning in STEM education research using R. The materials consist of four parts - what we called learning labs. Here are what each of the four are about:

= Learning Lab 1: Prediction — How is prediction different from explanation?
- Learning Lab 2: Interpretation — How do we interpret the accuracy of a machine learning model?
- Learning Lab 3: Feature Engineering — How can we improve our better?
- Learning Lab 4: Unsupervised methods — What if we do not have training data?

Each learning lab has two primary components: 1) slides, with a code-along portion (and code that participants and copy- and then paste/run along with the presentation) and 2) a more independent case study. There are also suggested readings in the case studies.

All of the materials are openly available here: https://github.com/laser-institute/machine-learning/tree/main; please use them and adapt them as you see fit.