A key early component of any machine learning project involves managing and understanding the data you have collected. Although you may not find it as gratifying as building and deploying models—the stages in which you begin to see the fruits of your labor—you cannot ignore the preparatory work.

Any learning algorithm is only as good as its input data, and in many cases, input data is complex, messy, and spread across multiple sources and formats. Because of this complexity, the largest portion of effort invested in machine learning projects is spent on the data preparation and exploration process.

This chapter is divided into three main sections. The first section discusses the basic data structures R uses to store data. You will become very familiar with these structures as you create and manipulate datasets. The second section is practical, as it covers several functions that are useful for getting data in and out of R. In the third section, methods for understanding data are illustrated throughout the process of exploring a real-world dataset.

By the end of this chapter, you will understand:

Since the way R thinks about data will define the way you think about data, it is helpful to understand the basic R data structures before jumping into data preparation. However, if you are already familiar with R data structures, feel free to skip ahead to the section on data preprocessing.