To calculate the overall star rating and percentage breakdown by star, we don’t use a simple average. Instead, our system considers things like how recent a review is and if the reviewer bought the item on Amazon. It also analyzes reviews to verify trustworthiness.
This is one of a handful of books on ML and R, and it’s by far the simplest in its coverage of content. For instance, it’s the only one I’ve looked at so far that has explained the difference between the Naive Bayes and Exact Bayes algorithms, although I haven’t looked at the Manning books on ML (e.g., Practical Data Science with R) some of which use R and some Python (e.g., Real-World Machine Learning). By a kindergarten approach to ML, I don’t mean to be disrespectful or to minimize the contribution of the book—presenting a topic at a simple basic level is worthwhile. The reasons I downgraded the book slightly were the large number of errors in the book (which look like material from previous editions that wasn’t caught) and some carelessness in what is presented. For some reason, in the chapter entitled “Overview of the Data Mining Process,” it bothered me a lot that the authors detail creation of dummy variables but don’t present factor variables until halfway through the book. Perhaps that was just the desire to present things at a basic level, however. I did learn some R functions I wasn’t familiar with from this book, and it does provide clear, explicit explanations of ML (or data mining) techniques. (For the life of me, I still don’t have a clear understanding of the differences between ML, data science, data analytics, and data mining—I suspect it is how the techniques and data are used that determines the difference, because the same techniques are used in all of these disciplines.) Hope this review was useful.
There are a lot of good things about this new book. First, this is a very good authoring team. They have a deep (both correct and very up-to-date) understanding that spans the technical aspects of data analytics, and how and why analytics is used in business settings to make business oriented decisions. Second, the book is well crafted. It is very well organized with very clear and to-the-point explanations. Of the five authors of this book, three of them (Shmueli, Bruce and Patel) have been together as a text writing team since 2007, and this is the 5th time they have updated and revised their business analytics text based on new developments in the field, and based on ongoing feedback from faculty and students who have used the text in both academic courses and professional training courses. Third, because this team has been deeply involved in data analytics for decades, they have a good sense of perspective. They understand the history and evolution of models from the statistics community, as well as the history and evolution of models from computing communities spanning both data mining and machine learning. The lead author, Galit Shmueli, has been publishing in both the statistics literature as well as the management science literature on the differences between explanation (and inference) versus prediction. This depth of perspective, and technically correct understanding of these important nuances, are reflected in how the material in this book is organized and presented. Fourth, they provide a unified and coherent approach to understanding these models. They have major sections on classification models and prediction models. They also have sections on time series forecasting, association models, and clustering. In each of these sections on model and application specifics, they draw on the appropriate models from across the communities of statistics, data mining and machine learning. In addition, the book contains sections on the front end (data prep, exploration, and dimension reduction), as well as on the backend of the modelling process (model evaluation). All examples are worked out in R.
The specific chapters on each data analytics modeling method are relatively short and to-the-point, as there are numerous textbooks and professional books on every one of the individual methods covered in this text. Because these authors are now in their fifth iteration of the content, and because they get a lot of feedback on what users of this material do and do not clearly understand, this authoring team has a knack for adding special explanatory material for those things that people tend to not understand well, or often misunderstand. While the chapters on each method are by design brief and introductory, they are solidly sufficient and highly informative, even for people with prior background in these methods. They have a good way of knowing what is important to explain, and a good way of explaining what they present.
In short, this is an excellent introductory text and also serves as a very good reference text for the most up-to-date thinking on the the modeling that underpins business analytics.
This book is very accessible for business professionals who want to develop their skills in data mining. I like how the "R" codes are presented and the ease in how it can be reproduced. Furthermore, each chapter is a stand-alone topic and does not generally require a sequenced order of reading from start to finish.
The book I received seemed to be in decent (new) condition, but had a few issues that seemed to be just poor quality. Every other page is connected (adhered) and had to be separated carefully without ripping.