Answer-first summary for fast verification
Answer: Model B is better than Model A because it has a lower FPR at the same TPR.
While the AUC is a useful metric for comparing the overall performance of classification models, it does not provide information about the trade-off between the true positive rate and the false positive rate at different thresholds. In this case, Model A has a higher AUC, indicating better overall performance, but it also has a higher false positive rate at the same true positive rate as Model B. This means that Model A is more likely to classify negative instances as positive, which may not be desirable in certain applications. Therefore, Model B is better than Model A in this specific scenario because it has a lower false positive rate at the same true positive rate.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
In a machine learning project, you are comparing the performance of two models using the Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) metric. Model A has an AUC of 0.92, while Model B has an AUC of 0.87. However, you notice that Model A has a higher false positive rate (FPR) at a given true positive rate (TPR) compared to Model B. Which of the following statements is true, and why?
A
Model A is better than Model B because it has a higher AUC.
B
Model B is better than Model A because it has a lower FPR at the same TPR.
C
Model A is better for imbalanced datasets, because it has a higher AUC.
D
Model B is better for balanced datasets, because it has a lower FPR at the same TPR.
No comments yet.