After preparing data and constructing meaningful features, the focus shifts to building models that can learn patterns and make predictions. This chapter concentrates on supervised learning, where models learn from labeled examples.

We will start with a brief review of common regression and classification algorithms before moving to their practical implementation using Python's scikit-learn library. You will apply linear models like Linear and Logistic Regression, tree-based methods such as Decision Trees and Random Forests, and ensemble techniques including Gradient Boosting Machines (XGBoost, LightGBM).

A significant part of building effective models involves evaluating them correctly and optimizing their settings. We will cover evaluation metrics suitable for different tasks, going beyond basic accuracy to include Precision, Recall, $F_1$ score, and ROC AUC. You will learn to implement robust validation strategies like $k$-fold cross-validation. Finally, we will address hyperparameter tuning, using systematic approaches like Grid Search and Randomized Search to find optimal model configurations.

By the end of this chapter, you will be able to train, rigorously evaluate, and tune several standard supervised learning models for predictive tasks.