AWS Certified AI PractitionerResponsible AIHard
A research team is developing an AI model to detect early signs of a rare disease. Due to the rarity, misdiagnosing a patient (false negative) could have severe, life-threatening consequences. Conversely, a false positive might lead to unnecessary further testing but is less critical than a missed diagnosis. Which evaluation metric should the team prioritize to ensure the model catches as many true positive cases as possible, even if it means a higher rate of false positives?
- AAccuracy
- BSpecificity
- CRecall (Sensitivity)
- DPrecision
Show answer & explanationAnswer & explanation
Correct answer: C. Recall (Sensitivity)
Recall (Sensitivity) measures the proportion of actual positive cases that are correctly identified. In a scenario where missing a positive case (false negative) is very costly, maximizing recall is paramount, even if it leads to more false positives (lower precision).
Why the other options are wrong
- A. Accuracy measures overall correct predictions, but doesn't differentiate between types of errors, which is critical here.
- B. Specificity measures the proportion of actual negative cases that are correctly identified, which is less critical than identifying true positives here.
- D. Precision measures the proportion of positive identifications that were actually correct, which would be low if false positives are high.
Recall (Sensitivity)
Recall (also known as Sensitivity) is the proportion of actual positive cases that are correctly identified by the model.
- Focuses on minimizing False Negatives (missing a positive).
- Calculated as True Positives / (True Positives + False Negatives).
- High recall is crucial in scenarios where missing a positive is very costly (e.g., disease detection).
Memory trick: Recall: Don't 'forget' any positive cases, 'recall' them all!