Machine learning is an interdisciplinary field that involves computer science, neurocomputing, statistics, and more. The idea of machines actually learning can be dated back to Alan Turing and the beginning of Artificial Intelligence (AI). Although the foundations of machine learning and the vague idea of it could be found earlier in the sayings of the great Turing, it was not until 1959 that the term machine learning, was coined by the computer scientist, Arthur Samuel.

Although such ideas were circulating before 20^st century, it only became popular in the first decades of the 21^st century; since then, its reputation has skyrocketed. There are many reasons for this having happened—machine learning is extremely useful—but I would mostly point to two different reasons.

First, there is data volume. Huge volumes of data are being produced every day, everywhere. To process all this information, a much more efficient and novel way of doing it was needed. Machine learning methods aimed to solve this problem. Some of their methods are data-hungry and practically each of them is able to handle linear and non-linear relations.

The second reason is feasibility. Algorithms and computing power have improved rapidly; thus, allowing machines to learn from large datasets in a reasonable time. This chapter is designed to introduce readers to the world of machine learning while estabilhing some paralallels with traditional statistics. The chapter also demonstrates how to practially fit several machine learning models through R.

Here is what can be found in this chapter:

• Which big companies are using machine learning
• Linear regression with base R
• Building decision trees with tree and rpart
• Random forest, bagging, and boosting methods
• Training support vector machines (SVM) with caret
• Building feedforward neural networks using h2o

There are several machine learning models already available for R users. In this chapter, quite a few of them will be discussed in a practical manner. But what is machine learning? There are many definitions. The next section is defining machine learning and briefly discussing its use.