Question 13
Domain 2: ML Model DevelopmentAn ML engineer needs to use an ML model to predict the price of apartments in a specific location. Which metric should the ML engineer use to evaluate the model's performance?
Correct answer: D
Explanation
Mean absolute error (MAE) is used for regression tasks like predicting apartment prices because it measures the average absolute difference between predicted and actual values. For a continuous numeric target, MAE shows how far off the model’s predictions are in the same units as the price.
Why each option is right or wrong
A. Accuracy
Accuracy measures correct class predictions, which fits classification rather than numeric price prediction.
B. Area Under the ROC Curve (AUC)
AUC evaluates ranking quality for binary classification, not error on continuous regression outputs.
C. F1 score
F1 score balances precision and recall for classification tasks, not continuous-value prediction.
D. Mean absolute error (MAE)
Apartment price prediction is a regression problem because the target is a continuous numeric value, so the evaluation metric must quantify prediction error in the same units as the output. Mean absolute error is defined as \(\text{MAE} = \frac{1}{n}\sum_{i=1}^{n}|y_i-\hat{y}_i|\), which directly measures the average absolute deviation between predicted and actual prices, making it appropriate for this use case.