Machine learning models have been around for a while. It is not surprising that machine learning has come to play an important role in data science.

Machine learning contributes to algorithms that can learn from, and make predictions on, data. Algorithms are built from example inputs into a machine learning model. Associating machine learning models and algorithms with big data has indicated the next stage of machine learning maturity. This book is a good introduction to machine learning models for big data classification, “which is one of the ... difficult problems in big data analytics.”

Typical of a Springer book, this one is concise, clear, and well organized. It introduces readers to many short coding examples in MATLAB, and R programming. Knowing both languages can help readers choose the right tasks to get the results they desire. For example, data from R can be exported into a MATLAB model the reader has built, and then the results can be passed back into R to create graphics.

After a brief introduction, the book spends the first part on understanding big data and the second part on big data systems using Hadoop and MapReduce as examples. Then, it moves on to the third part on machine learning concepts, “mainly modeling and algorithms; batch learning and online learning; and supervised learning (regression and classification) and unsupervised learning (clustering) using examples.” It shows how supervised learning models can be grouped into predictive and classification models, and looks at how supervised learning algorithms help to train the learning models.

The chapter on support vector machines provides readers with process diagrams and data flow diagrams to show how they can be implemented. The second part ends with a chapter on decision tree learning, which is a hierarchical supervised learning model.

The fourth part of the book focuses on scaling up machine learning. The first chapter introduces readers to the random forest supervised learning model that integrates a sampling technique with the decision tree model. The second chapter provides programming examples on three deep learning models: the no-drop, the dropout, and the dropconnect. It then moves on to the next chapter that proposes two new techniques: the chandelier decision tree and the random chandelier. The fourth chapter explains two dimensionality reduction techniques--feature hashing and principal component analysis--that can support scaling up machine learning.

As expected, the book comes with an appendix. Since each chapter contains programming examples and references, it is not necessary to have a separate part for them. Overall, this book is useful if you want to know more about machine learning models and algorithms for big data classification.