How to: Create a New OLAP Mining Structure
You can use the Data Mining Wizard in Microsoft SQL Server Analysis Services to create a new OLAP mining structure in an Analysis Services project or database.
To create a new OLAP mining structure
In Solution Explorer in Business Intelligence Development Studio, right-click the Mining Structures folder in an Analysis Services project, and then click New Mining Structure to open the Data Mining Wizard.
On the Welcome to the Data Mining Wizard page, click Next.
On the Select the Definition Method page, select From existing cube, and then click Next.
On the Create the Data Mining Structure page, select whether to create a mining structure only, or a mining structure plus one related mining model. If you will create a mining model, select the data mining algorithm that you want to use, and then click Next.
For more information about data mining algorithms, see Data Mining Algorithms (Analysis Services - Data Mining).
On the Select the Source Cube Dimension page, under Select a Source Cube Dimension, select the source cube dimension on which to base the mining structure, and then click Next.
On the Select the Case Key page, under Attributes, select the attribute that will be the key of the mining structure, and then click Next.
On the Select Case Level Columns page, under Related Attributes and Measures, select the attributes and measures that will be the case columns of the mining structure, and then click Next.
On the Specify Mining Model Column Usage page, under Mining model structure, select the input and predictable columns, and then click Next.
You can add a nested table to the mining structure using Add Nested Tables and remove nested tables using Remove Nested Tables.
If you add a nested table, specify the key of the nested table using Select Nested Table Key. Then, select the columns to include by using Select Nested Table Columns.
On the Specify Columns' Content and Data Type page, under Mining model structure, you can change the content type and data type for each column.
On the Slice Source Cube page, you can filter the data that is used to create the mining structure. For more information about how to slice a cube, see How to: Filter the Source Cube for a Mining Structure. Click Next.
On the Split data into training and testing sets page, specify a percentage of the mining structure data to reserve for testing, or specify the maximum number of test cases. Click Next.
If you specify both values, the limits are combined to use whichever is lowest.
On the Completing the Wizard page, provide a name for the mining structure and the related initial mining model that will be created.
If you want to create a mining model dimension, select the Create mining model dimension check box and provide a name for the dimension. The new dimension is saved in the cube used to build the structure.
If you want to create a new cube that contains the new mining model dimension, select the Create cube using mining model dimension check box, and provide a name for the new cube. The new cube contains both the existing dimensions that were used in building the structure, and the new data mining dimension that shows the results from the model.