on 18 December 2001
This is a welcome addition to the canon of books on data mining. It is an interdisciplinary book, drawing together the differing views of the statistician and the computer scientist, but with an emphasis on the principles underlying data mining. Many data mining books are written from a specialist computer scientist's viewpoint or from a similarly specialist business users one. In the former, the emphasis tends to be on algorithms and computational efficiency, while in the latter business applications of a small number of techniques are the main thrust.
Data mining requires an understanding of concepts from statistics and computer science, and the authors illustrate this with many examples. The first third of the book covers fundamentals of data analysis, which is appropriate, because some deep statistical ideas can arise in data mining problems, and those without training in statistics may not be aware of their consequences. The next third covers components of data mining algorithms, and the final part of the book draws the two strands together in a unified whole, with descriptions of typical data mining tasks and suitable algorithms.
It is a well written and easy to understand book, and will be an ideal reference for researchers and practitioners from either discipline, who may be seeking a greater understanding of the other.