ROC Tuning Steps

To perform ROC tuning:

1. Open the Properties pane for the Build node. Right-click the node and select Go to Properties.

2. In the Test section, select Generate Select Test Results for Model Tuning and run the node. See "Tuning Classification Models" for more information.

3. In the Models section, select the models that you want to tune and click .

4. Select Tune from the drop-down list. The Tune Settings dialog box opens in a new tab.

5. In the Tune Settings dialog box, go to the ROC tab.

6. If you are tuning more than one model, select a model from the Models list in the bottom pane. After you tune the first model, return to this pane and select another model.

7. Select a target value. In the case of ROC, there are only two values.

8. Select a custom operating point if you do no want to use the default point. See "Select Custom Operating Point" for more information.

9. Select the kind of Performance Matrix to use:

   • Overall Accuracy (default)

   • Average Accuracy

   • Custom Accuracy. Fill in the values for the Performance Matrix if you select this option.

   • Model Accuracy

10. Click Tune. New Tune settings are displayed in the same panel as the Performance Matrix. Examine the Derived Cost Matrix. You can continue tuning by changing any selections that you made.

11. When you have finished, click OK to accept the tuning, or click Cancel to cancel the tuning.

    • To reset the tuning, click Reset.

    • To see the impact of the tuning, run the Model node.