Preface
"It has been said that you don't really understand something until you have taught it to someone else. The truth is that you don't really understand it until you have taught it to a computer; that is, implemented it as an algorithm."
— Donald Knuth
As Don Knuth so wisely said, the best way to understand something is to implement it. This book will help you understand some of the most important algorithms in data science by showing you how to implement them in the Java programming language.
The algorithms and data management techniques presented here are often categorized under the general fields of data science, data analytics, predictive analytics, artificial intelligence, business intelligence, knowledge discovery, machine learning, data mining, and big data. We have included many that are relatively new, surprisingly powerful, and quite exciting. For example, the ID3 classification algorithm, the K-means and K-medoid clustering algorithms, Amazon's recommender system, and Google's PageRank algorithm have become ubiquitous in their effect on nearly everyone who uses electronic devices on the web.
We chose the Java programming language because it is the most widely used language and because of the reasons that make it so: it is available, free, everywhere; it is object-oriented; it has excellent support systems, such as powerful integrated development environments; its documentation system is efficient and very easy to use; and there is a multitude of open source libraries from third parties that support essentially all implementations that a data analyst is likely to use. It's no coincidence that systems such as MongoDB, which we study in Chapter 11, Big Data Analysis with Java, are themselves written in Java.