This book uniquely utilizes visualization tools to explain and study statistical learning methods. Covering classification and regression, the book is divided into two parts. First, various visualization methods are introduced and explained. Here, the reader is presented with applications of visualization techniques to learning samples (including projection pursuit, graphical matrices, and parallel coordinate plots) as well as functions, and sets. Next, the author provides a toolbox that contains formal definitions of the methods applied in the book and then proceeds to present visualizations of classified learning samples and classified test samples. Visualization methods are applied for the initial exploration of data, to identify the correct type of classifier, and to estimate the best achievable classification error. Once identified, the classifiers properties, proper uses, and overall performance are demonstrated and measured using visualization methods. Key areas of coverage include linear methods, kernel methods, additive models and trees, boosting, support vector machines, and nearest neighbor methods In addition to providing applications to engineering and biomedicine, the author also uses financial data sets as real data examples to illustrate nonparametric function estimation. The authors own R software is used throughout to reproduce and modify the books computations and research. Readers can duplicate these applications using the software, available via the books related Web site.