And why is it useful for BigData? 29 What is "continuous features binning"? What is a Standard Scaling? 38 Why are statistical distributions important? What is a Bias - Variance tradeoff? What is a training set, a validation set, a test set and a gold set in supervised and unsupervised learning? What is a cross-validation and what is an overfitting?

The material is arguably good, but the formatting is absolutely horrendous. The worst thing about this book is that it isn't written in Latex. It's probably written in MS Word, which most can agree has asinine handling of equations. Because of this, all of the equations and math syntax is offset from the rest of the text in each line, which makes it really distracting (and irritating) to read. Some of the tables have the header on one page and the data in the next -- again, a significant and egregious formatting issue that was just overlooked or flat-out ignored. Frankly I am surprised the author hasn't changed the formatting after so many different volumes.

And why is it useful for BigData? 29 Why are statistical distributions important? What is a training set, a validation set, a test set and a gold set in supervised and unsupervised learning? What is a cross-validation and what is an overfitting? Can you provide an example for Map and Reduce in Spark? What is a loss function, what are linear models, and what do we mean by regularization parameters in machine learning?

A nice balance of breadth and depth of you are already familiar with data science. If you are a data scientist this will give a nice review of topics that helps you pin point what you need to brush up on or quick primers to get you pointed in the right direction to study things you havent used much. I would strongly caution against using this book to learn a concept a new. Statistical concepts are given less weight and some are described with suboptimal accuracy. The programming/data engineering stuff is good and concise but maybe too brief for less CS oriented data scientists.

Myself as an interviewer, I would ask questions from this book. The Python examples are good and the code compact which shows great deal of effort. You will need to do your own Googling and research beyond the answers provided, but this book covers the field in sufficient breadth to give you confidence you will do well on an interview.

I had to comment to balance out the one single negative comment. It's true that there's some problems with the printing but the comment is irrelevant for the book's content. I think that the first volume is a great summary of subjects that junior data scientists must know. Also, I being a great believer in Occam's razor: if you managed to explain (or at least to remind) how does an SVM works in 2 pages instead of 20 pages - then you did a good job. The internet (and the library?) is full of details for further research.