The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition (Springer Series in Statistics) by Trevor Hastie (Author), Robert Tibshirani (Author), Jerome Friedman (Author). During the previous decade there was an explosion in computation and data technology. With it have come huge quantities of data in a wide range of fields comparable to medication, biology, finance, and marketing.
The challenge of understanding these information has led to the development of new instruments within the area of statistics, and spawned new areas akin to knowledge mining, machine studying, and bioinformatics. Many of those tools have frequent underpinnings however are often expressed with totally different terminology. This book describes the necessary ideas in these areas in a standard conceptual framework. Whereas the strategy is statistical, the emphasis is on concepts slightly than mathematics. Many examples are given, with a liberal use of colour graphics. It's a priceless useful resource for statisticians and anyone eager about knowledge mining in science or industry. The book's coverage is broad, from supervised studying (prediction) to unsupervised learning. The various topics include neural networks, assist vector machines, classification bushes and boosting---the primary comprehensive treatment of this matter in any book. This main new edition options many matters not covered within the original, together with graphical fashions, random forests, ensemble strategies, least angle regression & path algorithms for the lasso, non-unfavourable matrix factorization, and spectral clustering. There may be additionally a chapter on strategies for ``huge'' data (p greater than n), including a number of testing and false discovery rates.
his guide describes a lot of the important matters in machine learning. Most machine learning books simply present a criterion and and an optimization algorithm. For instance, LDA is usually introduced as: right here is the Fisher criterion, it seems like a great factor to maximize. "The Components of Statistical Studying" also presents that that is the precise criterion if the distributions of the information for every class are Gaussian with the identical covariance. This book puts all the algorithms in the identical statistical language, which makes them easy to check and choose between.
I also recognize the emphasis this book puts on algorithms which might be extra not too long ago fashionable/effective. I very a lot respect the discussions of logistic regression vs. LDA, ridge and lasso regression, boosting/additive logistic regression and additive trees, decision and regression bushes, ...
The one qualm I have with this ebook is that it's fairly biased toward the authors' personal research. It's tough from reading this ebook alone to differentiate between classical strategies and the authors' current proposed algorithms.
i really like this book. i have not completed reading yet. it is extraordinarily dense. by that, i imply every web page, each paragraph is packed stuffed with information. it makes for sluggish but very rewarding reading. i bought the book as a result of
i wanted to be taught one thing in regards to the topic. i've received a math and statistics background, however i haven't dealt with the broad topic of information mining or statistical learning. the e book fits my needs very very well.
it's clearly written. i have never found any grammatical or technical errors. it is pacing is ambitious, but i discover i can observe it. i do think some math and statistics background is required to make the e-book readable and useful.
i would not hesitate to recommend it to someone with the suitable background.
The Elements of Statistical Learning: Data Mining, Inference, and
Prediction, Second Edition (Springer Series in Statistics)
Trevor
Hastie (Author), Robert Tibshirani (Author), Jerome Friedman (Author)
768 pages
Springer; 0002-2009. Corr. 3rd edition (February 9, 2009)
No comments:
Post a Comment