Question 18
Domain 3: Model DevelopmentA data scientist wants to estimate how a model is likely to perform by repeatedly fitting it on different subsets of the same dataset and evaluating it across several partitions. Which model-fitting technique is being used?
Correct answer: B
Explanation
Cross-validation estimates model performance by fitting and evaluating the model across multiple folds of the dataset rather than relying on a single split. — official.txt
Why each option is right or wrong
A. Training the model once on the full dataset without partitions
Cross-validation evaluates performance across multiple folds, not a single fit on all data.
B. Using cross-validation during model fitting to assess performance over multiple folds
The scenario describes integrating cross-validation into model fitting by repeatedly fitting the model on different subsets and estimating performance across multiple folds, which matches the source material exactly.
C. Selecting features before fitting the model and skipping repeated evaluation
Feature selection does not by itself estimate performance across multiple folds.
D. Evaluating the model on one partition after a single training run
Cross-validation uses multiple folds rather than one partition after one training run.