Machine Learning With R — A Review
“Machine Learning with R” is an instructional tutorial which makes use of real-life cases to illustrate real-life application of machine learning techniques. It is written by Justin Lechter and Jason Lewis, with additional references from relevant journals and industry experts. The book has been developed for general purposes and may not be useful for scientific programming. However, students who have already learned linear algebra or calculus will find this text suitable for their intermediate or advanced courses in machine learning.
Machine learning with R presents its topics in a clear, concise style, which makes it easier to understand than texts dealing with more mathematically-acclaimed topics. The book contains twelve chapters, each dealing with a specific application area of Machine Learning. Each chapter consists of one application example, together with clear, detailed text explaining how the technique is used. Besides the application chapters, the book contains nine lab exercises and four practice sections. The exercises help students build up skills through simulated testing, and students are therefore encouraged to apply the concepts they learn in the labs.
Justin Lechter and Jason Lewis present a simple approach to the subject, and their book is therefore suitable for people already familiar with machine learning methods. A clear division between theory and practice helps readers follow the methods through experiments, while introducing them to the theoretical background first. The book is therefore well suited for graduate students in areas such as finance, information systems, computer science and more.
The main emphasis of the text is on supervised learning with a supervised data mining approach, thus covering both supervised and unsupervised learning. supervised learning deals with training and testing machines using labeled data, in which the accuracy of the results is dependent on the output label. supervised data mining applies the concept of supervised learning to unsupervised processes, where the accuracy of the output depends solely on the input data. It is therefore useful for training architectures and decision trees. The book thus covers both supervised and unsupervised data mining, with a focus on practical implementation and analysis.
Data Mining is an area of strong interest to machine learners, due to its wide applicability. The book provides detailed explanations and examples for creating decision trees and finding the maximum value from raw data. The book also introduces the concept of greedy data mining, and how it differs from traditional data mining methods, as well as why some applications might be better than others. The book briefly surveys previous literature on data mining, and reviews current approaches and their strengths and weaknesses. The authors conclude the book by briefly surveying the literature on data mining.
Machine Learning with R provides a solid foundation for machine learning practitioners, who may be new to the field or just interested in applying machine learning techniques to different domains. The book is very informative, and the primary focus of the book is on supervised learning and the application of supervised learning methods. Although the primary focus is on supervised learning, the book is helpful to machine learners and trainers in unsupervised learning as well, and the authors do not provide information on why one would use one method over the other. However, the book is very useful to anyone who is already using machine learning techniques, and is a good reference for experienced and new practitioners alike.
