Customer Reviews

Computer Systems That Learn: Classification and Prediction Methods from Statistics, Neural Nets, Machine Learning and Expert Systems (Machine Learning Series)

5 star:		(2)
4 star:		(1)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

This book gives a good cohesive introduction to the basic algorithms from Statistics, Machine Learning and Pattern Recognition research. These include Nearest Neighbor, Decision Trees, Bayesian Networks, and Neural Networks.

The main value of the book however is its coverage of techniques that 1) estimate a model's accuracy, and 2) select a 'good' model. This book offers the reader a solid foundation to what we are trying to achieve: to get at the objective truth.

I agree with another reviewer in the fact that this book contains many things that are omitted in the literature of this type - especially the basics, that are considered 'common sense' in the field. The authors of this book, however, realize that nothing is common sense in the scientific community, and the simplest of facts should be stated with perfect clarity, on several iterations, because otherwise even the basic principles will not be heeded.

This is exactly the approach the authors use - they build the knowledge from the ground up, which is great for someone like me, who only started working with neural networks (and classifiers in general) only several months ago. The book is perfect for someone who wants to get acquainted with classifiers (statistical, neural networks, etc.) I recommend it to anyone who did not cover this matter specifically, including (if not especially) those who consider themselves experts in the field of one classifier but not another - as the book gives a good overview of the used methods of classification, and gives a big picture with a bunch of references that help you better familiarize yourself with a particular topic in more depth.

This is the first book that made sense to me on the topic, and I strongly recommend it as the start-up guide for anyone new to the field.

This is a classic for anyone interested in machine learning, data mining or predictive statistics. Though it is beginning to age, it covers essential aspects of empirical modeling still not covered by many more recent titles (!). A subsequent effort by one of the authors, "Predictive Data Mining" is a bit more current though shorter on the fundamentals.