SQL Server Data Mining—Creating a Cluster Data Mining Model

Microsoft .NET Framework, ASP.NET, Visual C# (CSharp, C Sharp, C-Sharp) Developer Training, Visual Studio

Creating a Cluster Data Mining Model

We start the data mining process by creating a new mining model for the customer segmentation exercise, using an existing Analysis Services project that contains the cubes and dimensions with both the data warehouse information (such as in-store sales) and the new Internet information described earlier.

In Analysis Services data mining, we define a mining structure that describes the underlying data that will be used for data mining. Each mining structure can contain multiple mining models, such as a clustering model and an association model, that all use the same underlying data but in different ways.

The wizard will create the mining structure and model and open the structure in the data mining designer. The underlying data that you selected is shown on the Mining Structure tab, and the Customer Internet Segmentation model is the only model in the list on the Mining Models tab.

Before working with the model, you need to deploy the solution and process the mining model. During processing, Analysis Services applies the algorithm you selected (Microsoft Clustering) to the data from the cube to allocate all the customers to their appropriate clusters—your next task in data mining is to understand the information that has been produced and relate it to the real world.