Data Science Model Performance and Evaluation 3

Question 1

For a regression model, which metric penalizes large errors more heavily because it squares the residuals?

Accepted Answer

Mean Squared Error (MSE)

Answer

Mean Absolute Error (MAE)

Answer

Median Absolute Error

Answer

R-squared

Question 2

R-squared of 0.0 in a regression model indicates the model:

Accepted Answer

Explains no more variance than predicting the mean

Answer

Perfectly fits the data

Answer

Has negative residuals

Answer

Is overfitting

Question 3

When tuning hyperparameters, why should you use a separate validation set rather than the test set?

Accepted Answer

To avoid leaking test information and overly optimistic estimates

Answer

To increase training size

Answer

To reduce computation

Answer

Because test sets cannot be reused at all

Question 4

A model's predicted probabilities are systematically too confident. Which technique addresses this?

Accepted Answer

Probability calibration (e.g., Platt scaling)

Answer

Feature scaling

Answer

Increasing learning rate

Answer

Dropout

Question 5

In the bias-variance tradeoff, a model that is too simple to capture the underlying pattern exhibits:

Accepted Answer

High bias, low variance

Answer

High variance, low bias

Answer

Low bias, low variance

Answer

High variance, high bias

Question 6

Which metric is most appropriate when both false positives and false negatives carry costs and classes are imbalanced?

Accepted Answer

F1 score

Answer

Accuracy

Answer

True negative rate

Answer

Mean squared error

Question 7

A precision-recall curve is generally preferred over an ROC curve when:

Accepted Answer

The positive class is rare (highly imbalanced)

Answer

Classes are perfectly balanced

Answer

The model is a regressor

Answer

There are no false positives