Creating parameterized predictors

Pega Customer Decision Hub™ uses Adaptive Models that use a wide range of predictors to optimize customer interactions. You can make input fields that are not directly available in the customer Data Model accessible to the models by configuring these fields as parameterized predictors. Learn how to create parameterized predictors for Expressions, offline model scores, and online model scores.

This demo shows you how to create parameterized predictors to make input fields that are not directly available in the customer Data Model accessible to Adaptive Models. We will cover three examples: Expressions, offline model scores, and online model scores.

U+ Bank is cross-selling its credit cards on the web by using Pega Customer Decision Hub™. All customer data, including financial clickstream summary attributes, such as webpage visits, is available to the Adaptive Models that determine which offer to display for a particular customer.

The first parametrized predictor is the ratio of two clickstream summary attributes that denote the number of visits to the U+ Bank Investment webpage in the last 30 days and the last 90 days. An increasing ratio might indicate growing customer interest.

The Financial Services Clickstream Data Set contains the clickstream data, which is a record of the user's activity on the website. These entries reflect the Investment webpage visits of the customer.

To populate a parameterized predictor, you configure the NBA Pre-Processing Extension Point Decision strategy that is included in a change request. The pre-processing strategy runs before the Prediction strategy so that the output of the extension strategy is available as input to the Prediction.

The first example of a parameterized predictor is an Expression that references two clickstream fields. To add a field, you add a Set Property component to the strategy and create a new property in the Strategy Results class. Set the property type to Decimal to store the ratio of customer visits to a web page in the last 30 days and in the last 90 days.

This Expression returns a value of zero when the number of Investment page visits in the last 90 days is zero. Otherwise, it returns the ratio of the Investment page visits in the last 30 days to the last 90 days. A high value might indicate the increasing interest of the customer in the content of this page.

The second example of a parameterized predictor is offline model scores. A Data Set that is associated with the customer context makes the data available as a data source. Each customer can have multiple scores for different categories or different channels.

To add the model scores as potential predictors to the Adaptive Models, you add a Data Join component to the extension strategy.

The component joins the Offlinescores page that contains the model scores to the available data.

To output only relevant scores, you specify the channel and the product category to match the Prediction.

You then create a Decimal property and map it to accommodate the model scores.

The third example of a parameterized predictor is the score of a churn Prediction that runs in Customer Decision Hub. The Data Scientist team develops Predictive Models based on the latest insights and data that drive the churn Prediction.

To make the churn score available to the Adaptive Models as a parameterized predictor, add a Prediction component to the extension strategy. Reference the churn Prediction in the Customer class.

Add a Set Property component to the extension strategy to create a new property and map it to the propensity calculated by the churn Prediction.

Run the strategy by using a test customer profile to verify that the strategy outputs include the real-time churn risk score, the offline product scores, and the Expression.

Next, you use Prediction Studio to map three new parameterized predictors in the Prediction. The Predict Web Propensity Prediction calculates the likelihood that a customer clicks on a web banner. Like the new properties, the system saves the Prediction results to the CDH-SR class.

You then create three new parameterized predictors and map them to the new properties that represent customer behavior, offline scores, and online scores.

When you submit your work for deployment, the application saves all changes made in Predictions to a Change prediction change request.

After a review by the Team Lead, a Revision Manager deploys the changes as part of a revision.

You have reached the end of this video. You have learned:

• How to configure the pre-processing extension strategy for parameterized predictors.
• How to add properties as parameterized predictors to the Prediction.
• How to submit the changes for deployment.