Book Image

R Machine Learning By Example

By : Raghav Bali
Book Image

R Machine Learning By Example

By: Raghav Bali

Overview of this book

Data science and machine learning are some of the top buzzwords in the technical world today. From retail stores to Fortune 500 companies, everyone is working hard to making machine learning give them data-driven insights to grow their business. With powerful data manipulation features, machine learning packages, and an active developer community, R empowers users to build sophisticated machine learning systems to solve real-world data problems. This book takes you on a data-driven journey that starts with the very basics of R and machine learning and gradually builds upon the concepts to work on projects that tackle real-world problems. You’ll begin by getting an understanding of the core concepts and definitions required to appreciate machine learning algorithms and concepts. Building upon the basics, you will then work on three different projects to apply the concepts of machine learning, following current trends and cover major algorithms as well as popular R packages in detail. These projects have been neatly divided into six different chapters covering the worlds of e-commerce, finance, and social-media, which are at the very core of this data-driven revolution. Each of the projects will help you to understand, explore, visualize, and derive insights depending upon the domain and algorithms. Through this book, you will learn to apply the concepts of machine learning to deal with data-related problems and solve them using the powerful yet simple language, R.
Table of Contents (15 chapters)
R Machine Learning By Example
Credits
About the Authors
About the Reviewer
www.PacktPub.com
Preface
Index

Feature selection


The process of feature selection involves ranking variables or features according to their importance by training a predictive model using them and then trying to find out which variables were the most relevant features for that model. While each model often has its own set of important features, for classification we will use a random forest model here to try and figure out which variables might be of importance in general for classification-based predictions.

We perform feature selection for several reasons, which include:

  • Removing redundant or irrelevant features without too much information loss

  • Preventing overfitting of models by using too many features

  • Reducing variance of the model which is contributed from excess features

  • Reducing training time and converging time of models

  • Building simple and easy to interpret models

We will be using a recursive feature elimination algorithm for feature selection and an evaluation algorithm using a predictive model where we repeatedly...