Open Access

Julia Frederike Landsiedel

• This thesis aims to extend an existing Open Educational Resource (OER), which is available as a GitHub repository, and provide an organized introduction to basic machine learning (ML) concepts and algorithms. Further models, followed by structured metadata for each object, will be included while adhering to the contribution guidelines of the OER and following the CC license. The Machine-Learning-OER Basics repository intends to provide a wide range of benefits by enabling diverse users to apply and distribute machine learning algorithms. The goal of this digital collection is to fill the existing gap for instructional material on using machine learning in OER as well as make it easier to learn ML concepts effectively. These ML models are developed using the programming language Python and the library scikit-learn, among other standard libraries. Jupyter Notebook will make it straightforward for the user to explore the code. In order to apply the models to various practical scenarios, a non-specific data set is selected. This work is considered a solution approach in that it includes adding classification models. A performance comparison of the models is conducted. This comparative analysis evaluates the efficiency of each model. The examination includes various metrics for measurement. This work serves as a written extension, providing comprehensive background information on the algorithms utilized within the repositories and the performance comparison.