This excellent book and is exactly what the title says it is: an introduction to statistical learning with applications in R. It covers a wide range of statistical learning methods as well as the latest advances in nonlinear methods, such as generalized additive models, bagging, boosting, and support vector machines with nonlinear kernels, to name a few.

Each chapter is conveniently separated into two parts: the first part presents the theory and various methods, which are explained with the use of (usually a plethora of) examples, and the latter part, where the theory is put to practice with (mini) applications in R. Usually, one or more applications are presented that clearly demonstrate the strengths and weaknesses of the theory/method presented earlier in the chapter. Regarding the data, the book already comes equipped with three real-world datasets, which are used throughout the book. The datasets are available for download from the book’s website. Hence, everything you will need in order to try out the applications/examples is already contained in this book. This makes it ideal for instructors interested in using the book as a class textbook.

Overall, the examples are easy to follow and the explanations are clear. No prior programming knowledge of any language is required, although programming experience will definitely help in understanding the applications faster. In fact, when I started reading this book, I had never used R before; upon reading the book and completing the lab sections at the end of each chapter, I now feel quite comfortable in writing an R application to solve a particular machine learning problem.

I personally found chapter 9, “Support Vector Machines (SVM),” to be of particular interest. Although the explanation is very clear and all of the equations are clearly presented, I really liked the fact that the reader also has the option to visually inspect the result of each step/equation in the form of a graph or plot using R. This most definitely helps the reader to gain a better insight on the intrinsic details of this rather complex method. Moreover, the method is presented in gradually more complex ways: at first, SVM is applied for binary classification using linear boundaries, then for multiple classes using linear boundaries, and finally using nonlinear boundaries, which allows the reader to first become familiar with the concept and then move on to the more difficult scenarios.

This book has a very strong advantage that sets it well ahead of the competition when it comes to learning about machine learning: it covers all of the necessary details that one has to know in order to apply or implement a machine learning algorithm in a real-world problem. Hence, this book will definitely be of interest to readers from many fields, ranging from computer science to business administration and marketing.

More reviews about this item: Amazon, Goodreads