Model Iteration DY0-001 Practice Questions (Page 3)

Questions 11–13

1. 11application · medium

   A healthcare analytics team is developing a model to predict patient readmission risk. The dataset is highly imbalanced, with only 5% positive cases. The team trained a model and is evaluating its performance. They observe that the model achieves 95% accuracy but only 40% recall on the positive class. The hospital's goal is to identify as many high-risk patients as possible for intervention programs, even if it means flagging some patients who will not be readmitted. Which evaluation approach best aligns with the business objective?

   • AOptimize for accuracy, since 95% accuracy indicates the model is performing well overall
   • BUse precision as the primary metric, since false positives are costly in healthcare
   • CUse the F1-score, since it balances precision and recall equally
   • DUse recall as the primary metric, since the cost of missing a high-risk patient is higher than the cost of a false positive
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   A data scientist has trained a model and wants to validate its performance before deployment. Which method involves splitting the data into k subsets, training on k-1 subsets, and evaluating on the remaining subset, repeating this process k times?

   • AHoldout validation
   • BCross-validation
   • CStratified sampling
   • DBootstrap sampling
   Reveal Solution Discussion Select an answer first
3. 13foundation · easy

   A data scientist is evaluating a regression model that predicts house prices. Which metric directly measures the average magnitude of the errors in the same units as the target variable?

   • ARoot Mean Squared Error (RMSE)
   • BMean Squared Error (MSE)
   • CMean Absolute Error (MAE)
   • DR-squared (R²)
   Reveal Solution Discussion Select an answer first