R is the most widely used language for statistics, data mining, and machine learning. Besides the language, R is also the environment and the engine that executes the R code. You need to learn how to develop R programs, just as you need to learn any other programming language you intend to use.

Before going deeper into the R language, let's explain what the terms statistics, data mining, and machine learning mean. Statistics is the study and analysis of data collections, and the interpretation and presentation of the results of the analysis. Typically, you don't have all the population data, or census data, collected. You have to use samples—often survey samples. Data mining is again a set of powerful analysis techniques used on your data in order to discover patterns and rules that might improve your business. Machine learning is programming to use data to solve a given problem automatically. You can immediately see that all three definitions overlap. There is no big distinction between them; you can even use them as synonyms. Small differences are visible when you think of the users of each method. Statistics is a science, and the users are scientists. Data mining users are typically business people. Machine learning users are, as the name suggests, often machines. Nevertheless, in many cases, the same algorithms are used, so there is really a lot of overlapping among the three branches of applied mathematics in analytical applications.

Let's not get lost in these formal definitions and instead start with R immediately. You will learn about:

• Parts of open source R
• Basic description of the R language
• Obtaining help
• R core syntax elements
• R variables
• R vectors
• R packages

This chapter assumes that you are already familiar with SQL Server and R tools, including SQL Server Management Studio (SSMS), RStudio, or R Tools for Visual Studio, so you can start to write the code immediately.