Getting Started With SAS Enterprise Miner 5.2 (ISBN 1599940027)

Quick introduction:
For those who need to quickly learn SAS Enterprise Miner 5.2, this book is a good start, as it can be read and its sample projects completed in less than one day. Statisticians and applied mathematicians who have used SAS as a statistical analysis tool throughout the years and now must use it for database programming or data mining may find using SAS in this manner somewhat awkward (not due to the ability of SAS to do these tasks but due to their inexperience), but the Miner 5.2 GUI makes the learning experience much more palatable. The SAS authors have done a fine job of relating essential information on how to use Miner 5.2 without being bogged down in fine technical details. Those readers who intend to use Miner 5.2 for things such as the construction of neural networks have no doubt chosen it because of trust in its algorithms. Coding neural networks by hand is time-consuming, as is the validation of the resulting architecture, so having a tool like Miner 5.2 can assist in quickly obtaining a neural network or for doing rudimentary data mining. Miner 5.2 is flexible, but this reviewer has encountered problems that are very awkward to implement using this software, such as doing predictive time series analysis with feed-forward neural networks assuming an arbitrary lag. Data mining, which used to be thought of as a subfield of artificial intelligence, has become very popular in many different businesses in the last decade. The goal of data mining is to find patterns in large amounts of data, and various algorithms are used to do this, all of them expressing various biases towards particular notions of what constitutes a cluster or notions of data proximity. The authors expect the reader to be very familiar with data mining, and this expectation allows them to emphasize the Miner 5.2 application rather than spending time on the basic rudiments of data mining. Experts in data mining may find some of the dialog in the book trivial or sophomoric, but if they recognize that the book is designed to teach Miner 5.2 and not data mining, they should find the book helpful.