Modifying the Forecasting Model (Data Mining Tutorial)
The mining structure that you created in the previous task contains a single forecasting model. Before you process and explore the model, you have to change its structure slightly and modify one of its properties.
For More Information
You can change the mining structure by using the Mining Structure tab of Data Mining Designer. When you created the model with the Data Mining Wizard, you used only three columns: TimeIndex, ModelRegion, and Quantity. The Forecasting table also contains an Amount column, which you can use to forecast the amount of sales. With the Mining Structure tab, you can add this column from the data source view to the mining structure.
The Amount column now exists as part of the Forecasting mining structure.
Because you added a new column to the structure, you must define how the model will use the column. You can specify how the column will be used on the Mining Models tab of Data Mining Designer.
The Mining Models tab lists the columns that the mining structure contains in the Structure column of the grid, and lists the columns that the model contains in a column that has the name of the model, in this case Forecasting. Click the names of the columns or the name of the model to make modifications.
|In the Mining Models tab, you can also create new models based on the same structure, and you can adjust the algorithm and column properties for each model. You must process the model before these changes take effect.|
In the Forecasting mining model, the Amount column is used as an input column and is also used to forecast future sales. Therefore, you must set the properties of the column so that it can be used as both an input column and a predictable column.
The Amount column is now both an input column and a predictable column.
You can also change the properties of individual columns by selecting the column and opening the Properties window. To open the Properties window, right-click the column name, and then select Properties. If you change a property within the column for an individual model, you can change the properties only for that model. However, when you change a property within the Structure column, the change affects every model that is associated with the structure.
If you select the Forecasting column, you can change the properties and parameters that are associated with the model. The Microsoft Time Series algorithm contains several parameters that affect how a model is created. For more information about these parameters, see Microsoft Time Series Algorithm.
For this tutorial, you will adjust the value of the PERIODICITY_HINT parameter for the Forecasting model. This parameter gives the algorithm information about how frequently the data is repeated. The data in AdventureWorks DW is patterned on a monthly basis, and the periodicity is at the yearly level. Therefore, you will set the PERIODICITY_HINT parameter to 12, to indicate that a pattern repeats itself every year.
Now that the structure and parameters for the mining model are complete, you can process the model.
For More Information
On the Mining Model menu of BI Development Studio, select Process Mining Structure and All Models.
The Process Mining Structure - Forecasting dialog box opens.
The Process Progress dialog box opens to display information about model processing. Model processing may take some time, depending on your computer.
After processing is complete, click Close in the Process Progress and Process Mining Structure - Forecasting dialog boxes.