Question 34
UnclassifiedConcept drift specifically refers to a change in:
Correct answer: B
Explanation
Concept drift means the data-generating process has changed, so the mapping from inputs to outputs is no longer the same. It specifically refers to a change in "the relationship between input features and the target variable," which can reduce model accuracy over time.
Why each option is right or wrong
A. The number of features over time
B. The relationship between input features and the target variable
In machine learning terminology, concept drift is the change in the conditional distribution P(Y|X), meaning the target’s dependence on the input features has shifted over time. That is distinct from covariate shift, which affects only the input distribution P(X), so the exam’s correct option matches the formal definition used in standard ML texts and practice.
C. The deployment platform
D. The number of training samples