×

A first course in machine learning. 2nd edition. (English) Zbl 1373.68006

Chapman & Hall/CRC Machine Learning & Pattern Recognition Series. Boca Raton, FL: CRC Press (ISBN 978-1-4987-3848-4/hbk+ebook). xxix, 397 p. (2017).
In a nutshell, this book covers statistical techniques, mainly regression, and some ideas used in machine learning such as validation, \(k\)-nearest neighbors and clustering. Many traditional and important machine learning methods such as rule induction, decision tree generation, neural nets and genetic algorithms, to name a fee, are skipped altogether.
The first chapter of this book covers linear modelling (least squares and maximum likelihood approaches). The third and fourth chapters present the Bayesian approach, e.g., marginal likelihoods and Bayesian inference. Chapter 5 discusses classification: Bayesian classifiers, KNN, and support vector machines. In Chapter 6 the authors present clustering.
The remaining topics of the book are: principal component analysis and latent variable models (Chapter 7), Gaussian processes (Chapter 8), Markov chain Monte Carlo sampling (Chapter 9) and advanced mixture modelling (Chapter 10).
For the first edition see [Zbl 1342.68004].

MSC:

68-01 Introductory exposition (textbooks, tutorial papers, etc.) pertaining to computer science
62H30 Classification and discrimination; cluster analysis (statistical aspects)
68T05 Learning and adaptive systems in artificial intelligence

Citations:

Zbl 1342.68004

Software:

Octave; R; Matlab
PDFBibTeX XMLCite