Most jobs need reasoning—drawing conclusions based on available data—by a person or an automated system. This book’s framework of probabilistic graphical models provides a generic approach to this problem. The method is model-based, allowing for the creation of interpretable models that may then be changed by reasoning algorithms. These models can also be trained automatically from data, which means they can be utilised in situations when manually building a model is difficult or impossible. Because uncertainty is an unavoidable part of most real-world applications, the book focuses on probabilistic models, which make uncertainty explicit and enable more accurate models.