In stratified k-fold cross-validation for a classification problem, what makes it 'stratified'?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — stratified k-fold means each fold preserves the target class mix of the full dataset. You need that when classes are imbalanced, so you don't get a fold with no minority-class examples.
Full Explanation
The correct choice is that each fold preserves approximately the same proportion of each target class as the full dataset. Stratified k-fold ensures each fold reflects the overall class distribution, which matters for imbalanced targets because it avoids folds with no minority-class examples. The choice about folds sorted by predicted probability is wrong because that is unrelated to stratification. The choice about clustering data into k groups is wrong because that is clustering-based splitting, not stratified cross-validation. The choice about training on stratified random samples is wrong because that describes sampling, not the stratification mechanism in cross-validation.