Dr. Uday Kamath is a volcano of ideas. Every time he walked into my office, we had fruitful and animated discussions. I have been a professor of computer science at George Mason University (GMU) for 15 years, specializing in machine learning and data mining. I have known Uday for five years, first as a student in my data mining class, then as a colleague and co-author of papers and projects on large-scale machine learning. While a chief data scientist at BAE Systems Applied Intelligence, Uday earned his PhD in evolutionary computation and machine learning. As if having two high-demand jobs was not enough, Uday was unusually prolific, publishing extensively with four different people in the computer science faculty during his tenure at GMU, something you don't see very often. Given this pedigree, I am not surprised that less than four years since Uday's graduation with a PhD, I am writing the foreword for his book on mastering advanced machine learning techniques with Java. Uday's thirst for new stimulating challenges has struck again, resulting in this terrific book you now have in your hands.

This book is the product of his deep interest and knowledge in sound and well-grounded theory, and at the same time his keen grasp of the practical feasibility of proposed methodologies. Several books on machine learning and data analytics exist, but Uday's book closes a substantial gap—the one between theory and practice. It offers a comprehensive and systematic analysis of classic and advanced learning techniques, with a focus on their advantages and limitations, practical use and implementations. This book is a precious resource for practitioners of data science and analytics, as well as for undergraduate and graduate students keen to master practical and efficient implementations of machine learning techniques.

The book covers the classic techniques of machine learning, such as classification, clustering, dimensionality reduction, anomaly detection, semi-supervised learning, and active learning. It also covers advanced and recent topics, including learning with stream data, deep learning, and the challenges of learning with big data. Each chapter is dedicated to a topic and includes an illustrative case study, which covers state-of-the-art Java-based tools and software, and the entire knowledge discovery cycle: data collection, experimental design, modeling, results, and evaluation. Each chapter is self-contained, providing great flexibility of usage. The accompanying website provides the source code and data. This is truly a gem for both students and data analytics practitioners, who can experiment first-hand with the methods just learned or deepen their understanding of the methods by applying them to real-world scenarios.

As I was reading the various chapters of the book, I was reminded of the enthusiasm Uday has for learning and knowledge. He communicates the concepts described in the book with clarity and with the same passion. I am positive that you, as a reader, will feel the same. I will certainly keep this book as a personal resource for the courses I teach, and strongly recommend it to my students.

Dr. Carlotta Domeniconi

Associate Professor of Computer Science, George Mason University