Book Image

Mastering Machine Learning with R - Third Edition

By : Cory Lesmeister
Book Image

Mastering Machine Learning with R - Third Edition

By: Cory Lesmeister

Overview of this book

Given the growing popularity of the R-zerocost statistical programming environment, there has never been a better time to start applying ML to your data. This book will teach you advanced techniques in ML ,using? the latest code in R 3.5. You will delve into various complex features of supervised learning, unsupervised learning, and reinforcement learning algorithms to design efficient and powerful ML models. This newly updated edition is packed with fresh examples covering a range of tasks from different domains. Mastering Machine Learning with R starts by showing you how to quickly manipulate data and prepare it for analysis. You will explore simple and complex models and understand how to compare them. You’ll also learn to use the latest library support, such as TensorFlow and Keras-R, for performing advanced computations. Additionally, you’ll explore complex topics, such as natural language processing (NLP), time series analysis, and clustering, which will further refine your skills in developing applications. Each chapter will help you implement advanced ML algorithms using real-world examples. You’ll even be introduced to reinforcement learning, along with its various use cases and models. In the concluding chapters, you’ll get a glimpse into how some of these blackbox models can be diagnosed and understood. By the end of this book, you’ll be equipped with the skills to deploy ML techniques in your own projects or at work.
Table of Contents (16 chapters)

Random forest

Like our motivation with the use of the Gower metric in handling mixed, in fact, messy data, we can apply random forest in an unsupervised fashion. Selecting this method has a number of advantages:

  • Robust against outliers and highly skewed variables
  • No need to transform or scale the data
  • Handles mixed data (numeric and factors)
  • Can accommodate missing data
  • Can be used on data with a large number of variables; in fact, it can be used to eliminate useless features by examining variable importance
  • The dissimilarity matrix produced serves as an input to the other techniques discussed earlier (hierarchical, k-means, and PAM)

A couple of words of caution. It may take some trial and error to properly tune the random forest with respect to the number of variables sampled at each tree split (mtry = ? in the function) and the number of trees grown. Studies done show that...