This chapter focuses on validating and evaluating machine learning models’ performance. It introduces key metrics for evaluating model performance, such as accuracy, precision, sensitivity, specificity, and the F1 score, with examples highlighting their relevance in various practical contexts. The chapter explains the confusion matrix as a foundational tool for understanding model outcomes. It distinguishes between Type I and Type II errors, emphasizing how their relative importance varies depending on the context. The second part of the chapter introduces practical methods of model validation, such as cross-validation, the 0.632 bootstrap, and cost-sensitive learning, as well as tools to compare the performance of different models: the Receiver Operating Characteristic (ROC) curve and the Area Under Curve (AUC). The Python lab guides readers through training and testing a Naïve Bayes classifier to evaluate the likelihood of hotel booking cancellation based on reservation details. The goal is to evaluate model performance and learn model refining techniques.