Artificial intelligence / machine learning
Probably everyone consciously or unconsciously uses programs that are described by the term "artificial intelligence" (AI). These programs require a large amount of data, the values of which they store in a cleverly summarized form. This process is called "machine learning". Once the program has learned in this way, it can predict other data. This often seems "intelligent" to us. Is it really intelligent or is the background just so complicated that we don't understand it spontaneously?
Article MNU J. 6 (2024) (in German) with online supplements
article in English with Online Supplements
You can introduce machine learning for a linear classification using a manual paper-and-pencil method with a triangle. Starting from this, a corresponding program code can be created in just a few steps. The method is suitable for use in computer science lessons, but also in other subjects if you want to apply a simple machine learning method to suitable subject-specific didactic content.
The video (click on the image on the left) shows the sequence of the perceptron and Adaline learning algorithms for a student's data set. The two algorithms run one after the other.