Preface
Welcome to the world of data science. This book introduces machine learning with the same tools that data scientists use in practice. The regression chapters use R with the lessR package, and the decision tree, ensemble, and clustering chapters use Python with scikit-learn and XGBoost.
The book presents the two primary types of machine learning, introduced in the following chapters.
- Supervised learning, 1 Supervised Machine Learning, predicts a known target from a set of features.
- Unsupervised learning, 13 Unsupervised Learning, finds structure in the features when there is no target.
Most of the book develops supervised learning, beginning with linear regression and building toward more flexible classification methods.
- Regression: linear regression, inference, predictive accuracy, multiple regression, influence and outliers, preprocessing features, prediction from categories, and assumptions and transformations.
- Classification: logistic regression, decision trees, and ensemble models of decision trees, random forests and XGBoost.
- Unsupervised learning: cluster analysis with K-means.