PEGACPDS88V1 practice question 17 of 25

A large online store uses Pega Customer Decision Hub to smoothly adapt to changing customer behavior. Adaptive models help accomplish this business objective as the models learn from customer responses. Which statement about adaptive models is correct? s

Full set: 140 questions
Answer options
A: Adaptive models perform a binary model calculation Correct answer
B: Adaptive models require underlying predictive models
C: Adaptive models require a historical data set to start learning
D: Adaptive models perform a continuous model calculation
Correct Answer:
A. Adaptive models perform a binary model calculation
Explanation

Pega's adaptive models are primarily designed to predict customer behavior in response to an action, which is fundamentally a binary outcome (e.g., accept/reject, click/ignore, purchase/no purchase). The model calculates a propensity score, which is the probability of a positive outcome. This score is a key component in the Next-Best-Action arbitration formula. While the models learn and update continuously from new data, the core calculation for propensity is for a binary classification problem.

Why Incorrect

B. Adaptive models require underlying predictive models: This is incorrect. Adaptive models are self-contained and self-learning; they do not depend on or require pre-built, static predictive models to function.
C. Adaptive models require a historical data set to start learning: This is incorrect. A key advantage of adaptive models is their ability to start learning from real-time customer interactions without a large historical dataset, unlike traditional predictive models.
D. Adaptive models perform a continuous model calculation: This is less accurate. While adaptive models can be configured to predict a continuous value (e.g., spend amount), their primary and most common use in Customer Decision Hub is for binary classification to determine propensity. ---

References

1. Pega Documentation - About adaptive models:

"Adaptive models in Pega Platform™ are used to predict customer behavior

for example

the likelihood of a customer accepting an offer." This description of predicting the "likelihood" of an event (accept vs. not accept) directly corresponds to a binary model calculation.

Source: Pega Platform 8.8 Documentation

"About adaptive models".

2. Pega Academy - Decisioning Models (8.8) Course:

In the module "Predicting customer behavior using real-time data

" it is explained that adaptive models calculate propensity. "Propensity is the likelihood of a customer to exhibit a specific behavior

such as clicking a link or accepting an offer." The outcomes used as examples (Clicked/Impression

Accepted/Rejected) are all binary.

Source: Pega Academy

Certified Pega Decisioning Consultant (8.8) Mission

"Decisioning Models" course.

3. Pega Documentation - Next-Best-Action basics:

The Next-Best-Action arbitration formula is defined as Propensity x Context Weight x Action Value. The propensity

calculated by the adaptive model

represents the probability of a positive response

which is the output of a binary classification.

Source: Pega Platform 8.8 Documentation

"Next-Best-Action basics".

Scroll to Top