The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition (Springer Series in Statistics) [Hardcover]

Trevor Hastie (Author), Robert Tibshirani (Author), Jerome Friedman (Author)

Book Description

Publication Date: February 9, 2009 | ISBN-10: 0387848576 | ISBN-13: 978-0387848570 | Edition: 2nd ed. 2009. Corr. 7th printing 2013

During the past decade there has been an explosion in computation and information technology. With it have come vast amounts of data in a variety of fields such as medicine, biology, finance, and marketing. The challenge of understanding these data has led to the development of new tools in the field of statistics, and spawned new areas such as data mining, machine learning, and bioinformatics. Many of these tools have common underpinnings but are often expressed with different terminology. This book describes the important ideas in these areas in a common conceptual framework. While the approach is statistical, the emphasis is on concepts rather than mathematics. Many examples are given, with a liberal use of color graphics. It is a valuable resource for statisticians and anyone interested in data mining in science or industry. The book's coverage is broad, from supervised learning (prediction) to unsupervised learning. The many topics include neural networks, support vector machines, classification trees and boosting---the first comprehensive treatment of this topic in any book. This major new edition features many topics not covered in the original, including graphical models, random forests, ensemble methods, least angle regression & path algorithms for the lasso, non-negative matrix factorization, and spectral clustering. There is also a chapter on methods for ``wide'' data (p bigger than n), including multiple testing and false discovery rates.

Editorial Reviews

Review

From the reviews: "Like the first edition, the current one is a welcome edition to researchers and academicians equally…. Almost all of the chapters are revised.… The Material is nicely reorganized and repackaged, with the general layout being the same as that of the first edition.… If you bought the first edition, I suggest that you buy the second editon for maximum effect, and if you haven’t, then I still strongly recommend you have this book at your desk. Is it a good investment, statistically speaking!" (Book Review Editor, Technometrics, August 2009, VOL. 51, NO. 3) From the reviews of the second edition: "This second edition pays tribute to the many developments in recent years in this field, and new material was added to several existing chapters as well as four new chapters … were included. … These additions make this book worthwhile to obtain … . In general this is a well written book which gives a good overview on statistical learning and can be recommended to everyone interested in this field. The book is so comprehensive that it offers material for several courses." (Klaus Nordhausen, International Statistical Review, Vol. 77 (3), 2009) “The second edition … features about 200 pages of substantial new additions in the form of four new chapters, as well as various complements to existing chapters. … the book may also be of interest to a theoretically inclined reader looking for an entry point to the area and wanting to get an initial understanding of which mathematical issues are relevant in relation to practice. … this is a welcome update to an already fine book, which will surely reinforce its status as a reference.” (Gilles Blanchard, Mathematical Reviews, Issue 2012 d) “The book would be ideal for statistics graduate students … . This book really is the standard in the field, referenced in most papers and books on the subject, and it is easy to see why. The book is very well written, with informative graphics on almost every other page. It looks great and inviting. You can flip the book open to any page, read a sentence or two and be hooked for the next hour or so.” (Peter Rabinovitch, The Mathematical Association of America, May, 2012)

From the Back Cover

During the past decade there has been an explosion in computation and information technology. With it have come vast amounts of data in a variety of fields such as medicine, biology, finance, and marketing. The challenge of understanding these data has led to the development of new tools in the field of statistics, and spawned new areas such as data mining, machine learning, and bioinformatics. Many of these tools have common underpinnings but are often expressed with different terminology. This book describes the important ideas in these areas in a common conceptual framework. While the approach is statistical, the emphasis is on concepts rather than mathematics. Many examples are given, with a liberal use of color graphics. It is a valuable resource for statisticians and anyone interested in data mining in science or industry. The book's coverage is broad, from supervised learning (prediction) to unsupervised learning. The many topics include neural networks, support vector machines, classification trees and boosting---the first comprehensive treatment of this topic in any book. This major new edition features many topics not covered in the original, including graphical models, random forests, ensemble methods, least angle regression and path algorithms for the lasso, non-negative matrix factorization, and spectral clustering. There is also a chapter on methods for ``wide'' data (p bigger than n), including multiple testing and false discovery rates. Trevor Hastie, Robert Tibshirani, and Jerome Friedman are professors of statistics at Stanford University. They are prominent researchers in this area: Hastie and Tibshirani developed generalized additive models and wrote a popular book of that title. Hastie co-developed much of the statistical modeling software and environment in R/S-PLUS and invented principal curves and surfaces. Tibshirani proposed the lasso and is co-author of the very successful An Introduction to the Bootstrap. Friedman is the co-inventor of many data-mining tools including CART, MARS, projection pursuit and gradient boosting.

See all Editorial Reviews

Product Details

• Hardcover: 768 pages
• Publisher: Springer; 2nd ed. 2009. Corr. 7th printing 2013 edition (February 9, 2009)
• Language: English
• ISBN-10: 0387848576
• ISBN-13: 978-0387848570
• Product Dimensions: 6.1 x 1.5 x 9.2 inches
• Shipping Weight: 3.3 pounds (View shipping rates and policies)
• Average Customer Review: 3.9 out of 5 stars  See all reviews (49 customer reviews)
• Amazon Best Sellers Rank: #27,600 in Books (See Top 100 in Books)
  • #6 in Books > Professional & Technical > Professional Science > Biological Sciences > Bioinformatics
  • #9 in Books > Computers & Technology > Computer Science > Artificial Intelligence > Machine Learning
  • #12 in Books > Computers & Technology > Computer Science > Artificial Intelligence > Human Vision & Language Systems

Most Helpful Customer Reviews

90 of 91 people found the following review helpful

4.0 out of 5 stars The Elements of Statistical Learning November 12, 2004

By R. Krause

Format:Hardcover

The book is written by some of the biggest names currently in the field, and thus is written at a certain level, this isn't a fault of the book or the authers, but rather it was written for a specific audience. However I did find it odd when they would occassionally explain basic readily known notation, but later on assume the reader is familiar with what I would regard as advanced notation, or leave out quite a few steps in their mathematics assuming the reader understands what they did. This book covers a wide range of techniques ranging from the more traditional to the current, and for each topic presents an overview of the technique and provides adequate references for further exploration.

The reader should have a good underlying understanding of linear algebra, statistics and probability theory and also be familiar with the techniques presented here. This book was used in a graduate engineering data mining class, and most of us struggled greatly with the book. This book probably would have been more appropriate if this was a book to augment another text, or if this had not been the first time we had seen topics such as those presented, this being the book to explain neural networks, support vector machines and whatnot when you've never seen them before makes for a very bewildering experience, but once you find a few journal articles the techniques actually are fairly easy to understand.

The book does not explain how to implement using software any of the techniques, this is a topic left up to other books, such as Modern Applied Statistics with S by Ripley and Venerables, and only in their discussion about apriori for association rules did I see that they state a software package. It would have been nice if they would have given some insight into how they created some of the great graphics that punctuate the book, perhaps as additional material on the website.

A book that is more down to earth for engineers, albeit different in scope, would be Duda and Hart's Pattern Classification, which I believe are electrical engineers and written more from an engineering standpoint. In addition the Duda and Hard book gives a lot of applications-based problems and has an associated MATLAB handbook to walk readers through building many types of learners, while this book the end-of-chapter excercises are almost exclusively proofs and theoretical excercises. Not a fault of the book, but rather just a difference and depends on what the reader wants to get out of it.

Ultimately, even though it did prove to be a rather confusing book, I have learned a lot from it and will continue to go through it to learn even more from it as it does tend to become more lucid the more I go through it.

104 of 110 people found the following review helpful

5.0 out of 5 stars Useful book on data mining February 6, 2002

By frank lindemann

Format:Hardcover|Amazon Verified Purchase

I use data mining tools in my financial engineering and financial modeling work and I have found this book to be very useful. This book provides two crucial types of information. First, it provides enough theory to allow a potential user to understand the essential insights that motivate specific techniques and to evaluate the situations in which those technique are appropriate. Second, the book gives the exact algorithms to implement the various techniques.
While no book I have seen covers every data mining methodology available, this one has the strongest coverage I have seen in additive models, non-linear regression, and CART/MART (regression/classification trees). It also has very strong coverage in many other areas. I highly recommend it.

38 of 39 people found the following review helpful

5.0 out of 5 stars data mining from the viewpoint of statisticians January 23, 2008

By Michael R. Chernick

Format:Hardcover

Data mining is a field developed by computer scientists but many of its crucial elements are imbedded in important and subtle statistical concepts. Statisticians can play an important role in the development of this field but as was the case with artificial intelligence, expert systems and neural networks the statistical research community has been slow to respond. Hastie, Tibshirani and Friedman are changing this.
Friedman has been a major player in pattern recognition of high dimensional data, in tree classification, regularized discriminant analysis and multivariate adaptive regression splines. He has also done some exciting new research on boosting methods.

Hastie and Tibshirani invented additive models which are very general types of regression models. Tibshirani invented the lasso method and is a leader among the researchers on bootstrap. Hastie invented principal curves and surfaces.

These tools and the expertise of these authors make them naturals to contribute to advances in data mining. They come with great expertise and see data mining from the statistical perspective. They see it as part of a more general process of statistical learning from data.

The book is well written and illustrated with many pretty color graphs and figures. Color adds a dimension in pattern recognition and the authors exploit it in this book. It is really the first of its kind that treats data mining from a statistical perspective and is so comprehensive and up-to-date.

The important statistical tools that are covered in this book include under the category of supervised learning; regression, discriminant analysis, kernel methods, model assessment and selection, bootstrapping, maximum likelihood and Bayesian inference, additive models, classification and regression trees, multivariate adaptive regression splines, boosting, regularization methods, nearest neighbor classification, k means clustering algorithms and neural networks. These methods are illustrated using real problems.

Similarly under the category of unsupervised learning, clustering and association are covered. They cover the latest developments in principal components and principal curves, multidimensional scaling, factor analysis and projection pursuit.

This book is innovative and fresh. It is an important contribution that will become a classic. The level is between intermediate and advanced. Good for an advanced special topics course for graduate students in statistics. A comparable text is the text by Mannila, Hand and Smyth.

This book made effective use of color and maintained a competitive price. This had a major impact on publishers like Wiley that could not sell a book at this size and initial price. Wiley is still looking for a book comparable to this one that they can use to compete with Springer-Verlag. I know this information because I heard from the Wiley acquisitions editor that I worked with on my two books.