Understand the Problem by Understanding the Data

• Published in: Machine Learning in Python pp. 23 - 73
• Main Author: Bowles, Michael
• Format: Book Chapter
• Language: English
• Published: United States John Wiley & Sons, Incorporated 2015; John Wiley & Sons, Inc
• Subjects: machine learning data set; new data set; parallel coordinates plots; Python; quantile‐quantile plots
• ISBN: 1118961749; 9781118961742
• DOI: 10.1002/9781119183600.ch2

This chapter is about opening up your new data set so you can see what is inside, get an appreciation for what you will be able to do with the data, and start thinking about how you will approach model building with it. It has two purposes. One is to familiarize you with data sets that will be used later as examples of different types of problems to be solved using the algorithms. The other purpose is to demonstrate some of the tools available in Python for data exploration. The chapter uses a simple example to review some basic problem structure, nomenclature, and characteristics of a machine learning data set. After establishing some common language, the chapter goes one by one through several different types of function approximation problems. The chapter also introduces several visualization techniques, such as quantile‐quantile (Q‐Q) plots and parallel coordinates plots.