ISLR

ISLR refers to the book An Introduction to Statistical Learning with Applications in R. First published in 2013 by Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani, a second edition appeared in 2019. The text is an accessible introduction to statistical learning, designed to bridge theory and practice with practical R code. It covers a range of supervised learning algorithms (linear and logistic regression, regularization methods such as ridge and lasso, decision trees, random forests, boosting, k-nearest neighbors, and support vector machines) and unsupervised learning (principal component analysis and clustering), along with core topics in model assessment and selection (resampling methods, cross-validation, bootstrap).

The book emphasizes intuition, clear explanations, and real-data examples, complemented by exercises and datasets. It has become widely used in university courses and online programs as an entry point to data mining and machine learning, especially for readers with a statistics or data analysis background.

A companion website provides datasets and R code used in the book; an accompanying R package named ISLR contains many of these datasets for hands-on practice. The second edition expands and updates material, reflecting advances in the field and making statistical learning accessible to readers new to the subject. The work is published by Springer-Verlag and is regarded as a standard resource for beginners seeking practical understanding of statistical learning techniques in R.