Exams / AIF-C01 / ML lifecycle, MLOps and metrics / Question 2
AIF-C01 practice question: ML lifecycle, MLOps and metrics (2 of 4)
Fundamentals of AI and ML A free question from our original AIF-C01 practice exam. Try it, then open the answer to see why each option is right or wrong.
A hospital is testing a model that screens scans for a serious but treatable condition. Doctors say that missing a patient who has the condition is far worse than sending a healthy patient for a follow-up check. Which metric should the team focus on MOST?
Choose one answer.
- AA. Precision, because it measures how many flagged patients truly have the condition
- BB. Accuracy, because it counts every correct prediction equally
- CC. Recall, because it measures how many patients with the condition are found
- DD. Root mean squared error, because it penalises large mistakes
Show the answer and explanation
Answer: C
- A. Wrong: Precision matters when false positives are costly. Here, extra follow-up checks are the lesser problem.
- B. Wrong: If the condition is rare, a model can score highly on accuracy while still missing many patients who have it.
- C. Right: Correct. Recall is TP / (TP + FN). Maximising it minimises false negatives, which are the missed patients.
- D. Wrong: RMSE measures error for regression problems that predict numbers. This is a classification problem.
Explanation
Recall measures the share of actual positives the model finds, so it is the metric to prioritise when false negatives are costly, as in disease screening. Precision is the one to prioritise when false positives are costly. F1 balances the two. A model can trade one against the other by changing its decision threshold.
Study
Keep practising
Answer questions with a score at the end, or practise just this topic with the explanation after each one. No sign-up needed.