Testing a Filtered Model (Basic Data Mining Tutorial)

Now that you have determined that the TM_Decision_Tree model is the most accurate, you should evaluate the model in the context of the Adventure Works Cycles targeted mailing campaign. The Marketing department wants to know if there is a difference in the characteristics of male bike buyers and female bike buyers. This information will help them decide which magazines to use for advertising and which products to feature in their mailings. 

In this lesson, we will create a model that is filtered on gender. You can then easily make a copy of that model, and change just the filter condition to generate a new model based on a different gender.

For more information on filters, see Creating Filters for Mining Models (Analysis Services - Data Mining).

Filtering enables you to easily create models built on subsets of your data. The filter is applied only to the model and does not change the underlying data source. For information on applying filters to nested tables, see Intermediate Data Mining Tutorials (Analysis Services - Data Mining).

First you will make a copy of the TM_Decision_Tree model.

To copy the Decision Tree Model

  1. In Business Intelligence Development Studio, in Solution Explorer, select ASDataMining2008.

  2. Click the Mining Models tab.

  3. Right click the TM_Decision_Tree model, and select New Mining Model.

  4. In the Model name field, type TM_Decision_Tree_Male.

  5. Click OK.

Next, create a filter to select customers for the model based on their gender.

To create a case filter on a mining model

  1. Right-click the TM_Decision_Tree_Male mining model to open the shortcut menu.

    -- or --

    Select the model. On the Mining Model menu, select Set Model Filter.

  2. In the Model Filter dialog box, click the top row in the grid, in the Mining Structure Column text box.

    The drop-down list displays only the names of the columns in that table.

  3. In the Mining Structure Column text box, select Gender.

    The icon at the left side of the text box changes to indicate that the selected item is a table or a column.

  4. Click the Operator text box and select the equal (=) operator from the list.

  5. Click the Value text box, and type M.

  6. Click the next row in the grid.

  7. Click OK to close the Model Filter.

    The filter displays in the Properties window. Alternately, you can launch the Model Filter dialog from the Properties window.

  8. Repeat the above steps, but this time name the model TM_Decision_Tree_Female and type F in the Value text box.

You now have two new models displayed in the Mining Models tab.

Models cannot be used until they have been deployed and processed. For more information on processing models, see Processing Models in the Targeted Mailing Structure (Basic Data Mining Tutorial).

To process the filtered model

  1. Right-click the TM_Decision_Tree_Male model and select Process Mining Structure and all Models

  2. Click Run to process the new models.

  3. After processing is complete, click Close on both processing windows..

View the results and assess the accuracy of the filtered models in much the same way as you did for the previous three models. For more information, see:

Exploring the Decision Tree Model (Basic Data Mining Tutorial)

Testing Accuracy with Lift Charts (Basic Data Mining Tutorial)

To explore the filtered models

  1. Select the Mining Model Viewer tab in Data Mining Designer.

  2. In the Mining Model box, select TM_Decision_Tree_Male.

  3. Slide Show Level to 3.

  4. Change the Background value to 1.

  5. Place your cursor over the node labeled All to see the number of bike buyers versus non-bike buyers.

  6. Repeat steps 1 - 5 for TM_Decision_Tree_Female.

  7. Explore the results for the TM_Decision_Tree and the models filtered for gender. Compared to all bike buyers, male and female bike buyers share some of the same characteristics as the unfiltered bike buyers but all three have interesting differences as well. This is useful information that Adventure Works Cycles can use to develop their marketing campaign.

To test the lift of the filtered models

  1. Switch to the Mining Accuracy Chart tab in Data Mining Designer in Business Intelligence Development Studio and select the Input Selection tab.

  2. In the Select data set to be used for Accuracy Chart group box, select Use mining structure test cases.

  3. On the Input Selection tab of Data Mining Designer, under Select predictable mining model columns to show in the lift chart, select the checkbox for Synchronize Prediction Columns and Values.

  4. In the Predictable Column Name column, verify that Bike Buyer is selected for each model.

  5. In the Show column, select each of the models.

  6. In the Predict Value column, select 1.

  7. Select the Lift Chart tab to display the lift chart.

    You will now notice that all three Decision Tree models provide significant lift from the Random Guess model as well as outperforming the Clustering and Naive-Bayes models.

Community Additions