Initializing predictive model monitoring
Generate reports to help you verify whether DMSample predictive models accurately predict customer behavior. Sample reports are generated when you run the InitializePMMonitoring activity that simulates historical customer responses to model predictions.
Before you begin
Generate DMSample data. See Initializing DMSample data.
Populating predictive monitoring data
- In Dev Studio, search for the InitializePMMonitoring activity.
- In the top-right corner of the activity rule form, click Actions > Run.
- In the Run Activity window, in the noOfDaysToSimulate field, provide the number of days for which to simulate responses to predictive models.
The recommended number of days is 4. Depending on your system resources, you can increase this value. However, the reports might take a prolonged amount of time to generate.
- Click Run.
- Verify that the model reports are populated with data:
- In the navigation panel, click Records > Decision > Predictive Model.
- Open any model from the list, for example, Predict Churn.
- Click the Monitor tab.
- If no data is displayed, click Refresh data.
- Review the performance and analytical reports that are associated with the model that you selected.
The following figure demonstrates a number of reports that were generated for a churn prediction model over a period of four days:
Predictive Model reports
You have fully configured DMSample and you are ready to discover the features of the Next-Best-Action standard on Pega Platform™.
What to do next
Start exploring decision management in DMSample by performing one of the following actions:
To view the main process outline for this tutorial, see Getting started with DMSample.
Published February 1, 2019 — Updated March 21, 2019