Introduction to Machine Learning

Citation

(2010), "Introduction to Machine Learning", Kybernetes, Vol. 39 No. 8. https://doi.org/10.1108/k.2010.06739hae.001

Introduction to Machine Learning

Introduction to Machine Learning

Article Type: Book reports From: Kybernetes, Volume 39, Issue 8

Ethem Alpaydin2nd ed.The MIT PressCambridge, MAFebruary 2010584 pp. (172 illustrations)ISBN 978-0-262-01243-0$55.005/£40.95 (cloth)

Keywords: Cybernetics, Artificial intelligence, Systems theory, Signal processing, Statistics

This is a new edition of an introductory text in machine learning that the author believes provides a unified treatment of machine learning problems and solutions. It is published as a second edition in the MIT’s Adaptive Computation and Machine Learning series.

The author is Professor Ethem Alpaydin from the Department of Computer Engineering at Bogazici University, Istanbul, Turkey.

This is a topic which was pioneered by cybernetics enthusiasts who saw the field as an opportunity to apply cybernetics to learning systems. The aim of machine learning is to use a computer system (the machine) take example data or past experience to solve a given problem. This involves programming the computer system so that it could, for example, analyze sales data to predict customer behaviour so that a task can be completed using minimum resources, and exact knowledge from bioinformational data.

The publishers say that “the second edition is a comprehensive textbook on the subject, covering a broad range of topics not usually included in introductory machine learning texts”.

This unified treatment of learning problems and solutions discusses methods from a variety of fields, which include:

• •

  statistics pattern recognition;

• •

  neural networks;

• •

  artificial intelligence;

• •

  signal processing control; and

• •

  data mining.

The author explains all the learning algorithms so that the reader can link the equations given in the text to the actual computer programs.

Readers who are familiar with the first edition will wish to know what additional information is included. There are new chapters on kernal machines, graphical models and Bayesian estimation. There is also an expanded coverage of statistical tests provided in a chapter on design and analysis of machine learning experiments. There are case studies available from the web all with downloadable results for those who are teaching the subject as well as more exercises. The publishers say that all the chapters of the original edition have been revised and brought up-to-date.