Chapters are written on an elementary level for students and pratictioners of modern data analysis techniques. Written mainly as a text but expanded to cover topics of interest to researchers in statistics and computer science by subject matter experts. The last chapter on Systems and Applications by Xiaohui Liu includes coverage of data quality. Among the references on data quality and outlier detection is the book edited by Wright "Statistical Methods and the Improvement of Data Quality". That book was a collection of papers from a conference held in Oak Ridge Tennessee in 1982. That volume was published by Academic Press in 1983. It is not often sighted in the statistical literature but it did contain a number of interesting papers. I contributed a chapter on influence function methods for outlier detection to the Academic Press book.

Hand has written many books on statistics and especially some excellent texts on classification and pattern recognition. His recent work on data mining was published in 1999 by MIT press, a volume he coauthored with Mannila and Smyth. it is one of teh few data mining texts that is highly regarded by the statistical community. Much of that work in referenced in this book particularly in Chapter 1, the overview chapter on intellegent data analysis that Hand wrote himself.

Resampling methods, generalized linear models, Bayesian methods, time series, multivariate analysis, random effects models and entropy are all covered with nice elementary introductions.

This is a great reference source with over 440 articles and books in the list of references.

The first part of this book is focused on classical statistical issues. Arguably, anyone seeking to perform advanced data analysis should have a working knowledge of this area. It is my personal observation that, unfortunately, many workers do not. This book provides a good way of gaining a broad understanding of statistical methods. My only caveat is that the discussion of naïve Bayesian classifiers could have been more extensive. (The chapter on general Bayesian classifiers is other wise well done.) Naïve Bayesian classifiers have been reasonably successful in machine learning and a more in depth treatment would have been useful.

The later chapters focus on machine learning. They provide useful introductions into: induction, neural networks, fuzzy logic, and stochastic search. These chapters are particularly useful to workers contemplating how to best perform advanced analysis of complex, large, and possibly imprecise data sets. Consequently, someone contemplating data mining or other intelligent data analysis applications should seriously consider acquiring this book.