Exploring Data in Engineering, the Sciences, and Medicine

This is a wonderful extended discussion of the intersection of traditional statistics with "big data". There's a lot of math and the mathematical development, as well as the choice of topics, is informed by the issues raised in data exploration, and therefore well motivated for readers like myself who are more engineering focused. The author is a bit obsessed with outliers and datasets that are corrupted in one way or another and analyzes several different kinds of examples throughout the book. Such an obsession is appropriate for a book about exploring data, and also for the context in which people are analyzing datasets created by others, so that the analysts aren't familiar with the measurement conditions, post-measurement processing, unstated assumptions, and so on. I found this book to be an efficient way to get a deeper understanding of statistics as it applies to practical data analysis. What I'm looking for now is a book teaches the relationships between the techniques of machine learning and traditional statistics.

This is definitely a classic and a reference book to have. However, it is not integrated into any programming language such as R. So it is a textbook in then best tradition.

In terms of clarity, breadth, and accuracy, I found this book to be way above the standards in its class. It is outstanding as a reference book and as a potential textbook.

I found the book to be well written. The examples are interesting. It is a nice balance of rigor and practicality.