Data Science Model Performance and Evaluation 2

Question 1

A spam classifier flags 95 of 100 actual spam emails but also flags 40 legitimate emails as spam. Which metric most directly reflects the cost of those false alarms?

Accepted Answer

Precision

Answer

Recall

Answer

Accuracy

Answer

Sensitivity

Question 2

On a dataset where 99% of cases are negative, a model that always predicts 'negative' achieves 99% accuracy. What does this illustrate?

Accepted Answer

The accuracy paradox on imbalanced data

Answer

Overfitting

Answer

High variance

Answer

Data leakage

Question 3

Which metric is the harmonic mean of precision and recall?

Accepted Answer

F1 score

Answer

ROC-AUC

Answer

Log loss

Answer

R-squared

Question 4

A regression model has a low training error but a much higher test error. This is a sign of:

Accepted Answer

Overfitting

Answer

Underfitting

Answer

High bias

Answer

Class imbalance

Question 5

The area under the ROC curve (AUC) represents the probability that the model ranks a randomly chosen positive higher than a randomly chosen negative. An AUC of 0.5 means:

Accepted Answer

No better than random guessing

Answer

Perfect classifier

Answer

Worse than random

Answer

Perfectly calibrated

Question 6

Which evaluation approach gives a more reliable performance estimate on small datasets by averaging across multiple train/test splits?

Accepted Answer

k-fold cross-validation

Answer

Single holdout split

Answer

Using all data for training

Answer

Bootstrap aggregation of features

Question 7

In a confusion matrix, recall (sensitivity) is calculated as:

Accepted Answer

TP / (TP + FN)

Answer

TP / (TP + FP)

Answer

TN / (TN + FP)

Answer

(TP + TN) / Total