Chapter 1. Introduction to Practical Machine Learning Using Python

In the technology industry, the skill of analyzing and mining commercial data is becoming more and more important. All the companies that are related to the online world generate data that can be exploited to improve their business, or can be sold to other companies. This huge amount of information, which can be commercially useful, needs to be restructured and analyzed using the expertise of data science (or data mining) professionals. Data science employs techniques known as machine learning algorithms to transform the data in models, which are able to predict the behavior of certain entities that are highly considered by the business environment. This book is about these algorithms and techniques that are so crucial in today's technology business world, and how to efficiently deploy them in a real commercial environment. You will learn the most relevant machine-learning techniques and will have the chance to employ them in a series of exercises and applications designed to enhance commercial awareness and, with the skills learned in this book, these can be used in your professional experience. You are expected to already be familiar with the Python programming language, linear algebra, and statistics methodologies to fully acquire the topics discussed in this book.

• There are many tutorials and classes available online on these subjects, but we recommend you read the official Python documentation (https://docs.python.org/), the books Elementary Statistics by A. Bluman and Statistical Inference by G. Casella and R. L. Berger to understand the statistical main concepts and methods and Linear Algebra and Its Applications by G. Strang to learn about linear algebra.

The purpose of this introductory chapter is to familiarize you with the more advanced libraries and tools used by machine-learning professionals in Python, such as NumPy, pandas, and matplotlib, which will help you to grasp the necessary technical knowledge to implement the techniques presented in the following chapters. Before continuing with the tutorials and description of the libraries used in this book, we would like to clarify the main concepts of the machine-learning field, and give a practical example of how a machine-learning algorithm can predict useful information in a real context.