Features
- Cover Type: Paperback with 400 pages
- Published by: Microsoft Press
- Edition: 1st Edition June 9, 2001
- Written in: English
- ISBN 10 Number: 0735612714
- ISBN 13 Number: 978-0735612716
-
Book Dimensions:
9.6 x 7.6 x 1.3 inches
- Weighs: 5.3 ounces
Product Review
Your organizational database is only as good as the strategic data you can extract from it. Do customers who buy breakfast cereal typically buy bananas as well? Is there a correlation between rainfall in a particular region and the prevalence of a particular illness there?
Data Mining with Microsoft SQL Server 2000 Technical Reference shows how to use
Microsoft's analysis tools for large databases. Author Claude Seidman offers advice on the data-modeling engineering process as a whole, including designing strategies likely to yield meaningful results, designing data warehouses, growing decision trees, spotting clusters and anomalies in data, and automating mining processes with code.
Despite its designation as a reference, this book is largely a tutorial--you'll refer to it for advice on how to make Analysis Services do something in particular. Seidman uses a classic and effective tutorial technique, sticking with an example throughout the book and adding to previous examples as he explores additional aspects of
Microsoft data mining. His illustration involves identifying edible mushrooms, based on a database of facts about known mushrooms, and he's combined how-to prose with screen shots and accumulated wisdom to great effect. If your organization has gone with
Microsoft SQL Server 2000 for data storage, read this book for advice on knowledge extraction.
--David Wall Topics covered: Microsoft Analysis Services, including the proper use of Data Transformation Services (DTS), PivotTable Services, Decision Support Objects (DSO), and the
Microsoft implementation of On-Line Analytical Processing (OLAP).
Product Description
With its state-of-the-art capabilities for rapidly processing and retrieving huge quantities of data,
Microsoft® SQL Server 2000 is quickly growing in popularity among large corporations. But learning how to take advantage of the powerful, built-in data-mining services in SQL Server to turn all that data into meaningful information takes time and effort. Data Mining with SQL Server 2000 Technical Reference is the ideal, in-depth reference guide for any database developer, administrator, or IT professional who requirements comprehensive information about these powerful new data-mining services. In particular, it fully looks at the data-warehousing architecture in SQL Server 2000 to show how to take full advantage of the data-mining services in this RDBMS. This is the only
Microsoft-approved technical guide to the data mining services in SQL Server 2000.
Reader ReviewsI have to agree with one of the previous reviewers when he said that given the absence of practically *ANY* documentation provided by Microsoft, this book is your only real source of information about Microsoft's data mining product. I'm a big fan of OLAP amd data mining which made me better appreciate the time the author took to lay the groundwork for the discipline of data mining. Unlike a previous reviewer, I think that the author shares lots of real-world evperience which you can see by the way he bring up problems (which I have encountered myself) that occur when moving from raw data to a data mining model. He also catches some glitches and unreported features in the product for you and shows you how to work around them. The book is actually very complete considering that the data mining product put out by Microsoft is promising, but extremely rudimentary. It provides only two basic data mining algorithms and gives a very clumsy way to try to add other algorithms. Thankfully, the author discusses techniques and pitfalls of mining numerical data and even shows you how to use SQL Server 2000 to perform a regression analysis for that purpose. I would have given this book five stars except for two points : 1: The mushroom database is a good illustration of the use of the decision tree algorithm, but I think it may have been good to include a more business-oriented example that would bring data mining closer to it's intended purpose. 2: I was a little disappointed not to see any explanation as to how to add your own algorithms to the data mining product. Even if doing so requires C++ experience, it would have been perfectly fine to include it in a separate chapter or in an appendix. I don't know why the author chose not to include it. Byond that, I would definitely recommend this book if you need to use MS data mining. The book is well written, and considering the infancy of the product, it's also very complete. Besides, you have no other real resource out there!
Search for books at
