Beginners Guide

Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations (The Morgan Kaufmann Series in Data Management Systems)

Format: Paperback
Author: Ian H. Witten
ReleaseDate: 11 October, 1999
Publisher: Morgan Kaufmann
Rating:

An Excellent Data Mining Text
It is written for practitioners and provides clear explanation of included topics. This book covers data mining at a serious level, including essential material on testing and a broad array of techniques. Easily one of the best 5 books on data mining currently available.

Note that this book has moved on to a second edition.

A good book to practice

With the software that you can dowload you can do yourself all the exercices for every models presented
It's the best way to progress
Do the same, it's simple and funny
The explanations are very clear and pedagogical, very practical. I have bought this book as course book to learn some particular aspects of data mining.

A nice complement to the other data mining bible
One could hardly do better than to own this book and "The Elements of Statistical Learning; Data Mining, Inference, and Prediction" by Hastie, Tibshirani, and Friedman which covers complimentary material from the statistician's perspective. Witten's book, combined with the accompanying open source package, Weka, provides a great overview of data mining principles and practice from a machine learning perspective. Witten does an amazing job of providing a comprehensive overview of the field while still providing some depth re. the algorithms; after reading the book I didn't feel like I'd read yet another large volume of empty claims about the power of information technology to make me rich and famous. In fact, with the book by my side, in a relatively short time I was able to use Weka to pry some useful information from one of my medical imaging data sets (maybe even enough to serve as preliminary data for a grant application). It seems to me that with an understanding of the material in this book and the one by Hastie et. al. one could embark on serious data mining projects. .