Modelling is the stage where the AI system is actually created. In this step, a suitable algorithm or model type is selected, such as regression, decision trees, or neural networks. The model is trained on the prepared data to learn patterns and relationships between inputs and outputs. Modelling is both a technical and experimental step: different algorithms may be tested, parameters adjusted, and techniques like feature selection or dimensionality reduction applied to improve performance.
Using the school example, a machine learning model can be trained to predict exam results based on past student data like attendance, homework, and prior test scores. The model learns patterns, such as students with low attendance and incomplete homework being at higher risk of failing.