MACHINE LEARNING (Int'l Ed) Paperback – 16 Oct. 1997

Best introductory book I have read on the topic. Emphasis is more on explaining what the algorithms do as opposed to providing many recipes with no real insight.

Structure very good. It is very useful for self-study.

The book covers a broad range of topics and approaches in machine learning. As a consequence, the amount of content dedicated to each topic is quite sparse. Decision Trees, Neural Networks, Bayesian Classifiers/Networks, Instance-Based Learning and Genetic Algorithms are all covered in a single book that counts under 400 pages. Since it is written in a concise and intuitive way however, it provides a solid foundation that the reader can build upon if he wishes to go deeper into any subject. Likewise, with this foundation, readers should be able to easily catch up on recent innovations (the book is quite old). Recommended.

This book provides a smooth introduction to Machine Learning. It is not too math heavy and can be used easily by people with math cs background. There little golden nuggets of concentrated experience scattered around which makes it even more worthwhile for people just diving in. Each chapter is independent and straight to the point. I highly recommend it

Covers Machine Learning concepts thoroughly, allowing you to decide which type is best for any particular problem. Uses some daunting mathematical notation, but still easy to follow