Importing a PMML model

Archived

2 Tasks

25 mins

Pega Customer Decision Hub 8.4 Visible to: All users

Beginner Pega Customer Decision Hub 8.4 English

Scenario

U+ Bank has recently implemented Pega Decision Management but has already been using predictive models created by an external bureau for some time. You have been asked to showcase an existing model that predicts churn in a new decision strategy.

Use the following credentials to log in to the exercise system:

Role	Username	Password
Data Scientist	DataScientist	rules

Your assignment consists of the following tasks:

Task 1: Import a PMML model

In Prediction Studio, create a predictive model, ChurnPMML, by using the following Predictive Model Markup Language (PMML) file: DecisionTree_ChurnApplySet.pmml. Map all the predictors and select Predicted_ChurnApplySetChurn for the result.

Task 2: Test the PMML model in a strategy

Modify the PMMLRetention strategy to showcase the PMML model. Set the model output Probability_Y to ProbabilityToChurn. Test the strategy using the Steve and Bill data transforms. Examine the results.

Challenge Walkthrough

Detailed Tasks

1 Import a PMML model

1. Download the DecisionTree_ChurnApplySet.pmml file.
2. Log in as a Data Scientist with user name DataScientist and password rules.
3. In the navigation pane on the left, click Intelligence > Prediction Studio.
4. Click New > Predictive Model.
5. In the New predictive model dialog box, enter the following information:
   1. Name: ChurnPMML.
   2. Click Import PMML.
   3. Click Choose File, and then select the DecisionTree_ChurnApplySet.pmml file.
6. Click Next.
7. In the Outcome definition section, enter the following information:
   1. Expected performance: 70. The expected performance is used as benchmark when monitoring the model.
   2. Monitor the probability of: Y
8. Click Import to import the PMML file and close the dialog box.
9. In Prediction Studio, click the Mapping tab.
10. In the mapping fields, enter the following information to map all the unmapped predictors to the relevant properties. For example, CUSTOMER_ID -> .CustomerID, RESBIZ -> .ResidentBusiness, AVG_AGE -> .AverageAge etc.
    
11. Click Save to save the model.

2 Import a PMML model

1. Click Back to Customer Decision Hub.
2. In the navigation pane on the left, click Intelligence > Strategies and open the PMMLRetention strategy.
3. Check out the strategy.
4. On the strategy canvas, right-click, and then select Decision Analytics > Predictive Model.
5. Right-click the new Predictive Model component, and then select Properties.
   1. In the Predictive model field, select ChurnPMML; the name self-populates.
   2. In the Output mapping tab, click Add Item.
   3. In the Target field, enter or select .ProbabilityToChurn.
   4. In the Source list, select the Probability_Y model output.
   5. Click Submit.
6. On the canvas, right-click, and then select Arbitration > Filter.
   1. Right-click the new Filter component, and then select Properties.
   2. Enter Churners as the name.
   3. To the right of the Filter condition field, click the Gear icon.
   4. In the Expression builder, enter ChurnPMML.pxSegment=“Y”.
   5. Click Submit to close the Expression builder.
   6. Click Submit to close the Filter properties.
7. Connect the Proposition Data component to the Filter component and the Filter component to the Results. The resulting strategy should look like the following image:
   
8. On the right, open the Test run pane, and then click Settings.
9. In the Data transform field, enter or select Steve.
10. click Save & Run.
11. In the Show component level values list, select ProbabilityToChurn; notice the high value.
12. In the Show component level values list, select Name and confirm the retention action is taken.
13. Repeat with the Bill data transform and confirm that no retention action is taken.

 

---

---

Available in the following mission:

• Data Scientist v1