The Microsoft® Data Warehouse Toolkit: With SQL Server™ 2005 and the Microsoft® Business Intelligence Toolset

10.5. Data Mining Examples

For many of us, the best way to learn something is by doing it. This is especially true for technical folks, and even more so for data mining. You really need to work through the SQL 2005 Data Mining Tutorial before you run through this example. This will give you a chance to get familiar with the tools and the user interface, which will make these examples easier to understand and follow. If you haven't already done so, please take the time to work through the data mining tutorials now.

In this section we start out with a simple example to get a feel for the data mining tool environment and the data mining process. In fact, the first example is so simple that its case set of economic data can be presented on a single page. Then we dig into a more detailed example based on the Adventure Works Cycles data mining tutorial data. Both of these examples will follow the general flow of the data mining process presented in the previous section.

10.5.1. Case Study: Classifying Cities

This first example is a small, simplified problem designed to provide a clear understanding of the process. The dataset has only 48 cases total—not nearly enough to build a robust data mining model. However, the small size allows us to examine the inputs and outputs to see if the resulting model makes sense.

The scenario is based on a large non-governmental organization with a mission and operations much like that of the World Bank:

(T)o fight poverty and improve the living standards ...