Acceptance criteria · Question 40 of 60
Which acceptance criteria are examples for AI functional correctness?
2 answers are correct. Medium
Show answer
Answer
- A Recall of 90% for a defect prediction system
- B Accuracy of 95% for an image recognition system
Explanation
Typical criteria for AI functional correctness are an accuracy of 95% for an image recognition system and a recall of 90% for a defect prediction system. Such thresholds define which error rate is acceptable. The response time during a database outage is a robustness criterion, and shutting down the production line one for intervenability.