CT-AI: Certified Tester AI Testing Version 2.0

By ISTQB · Question Mode

QuestionQ53

ML Functional Performance Metrics

“Splendid Healthcare” has begun developing an ML-based cancer detection system. The cancer type it intends to detect has a 2% prevalence rate among the population in a particular geography. The model must perform well for both normal patients and cancer patients.

Which ONE of the following combinations of functional performance metrics should be the MOST IMPORTANT target for improvement?

Explanation

For a condition with only 2% prevalence, accuracy can appear high by predicting most patients as normal. Recall is essential for identifying actual cancer cases, and precision is essential for ensuring cancer predictions are not dominated by false positives. Together, these metrics target meaningful performance for cancer detection under severe class imbalance.

Community Discussion

No comments yet. Be the first to start the discussion!