Evaluate a fraud classifier.
Reduce missed fraud while also tracking false alarms.
Choose two correct statements.
Precision is the fraction of positive predictions that are truly positive; recall is the fraction of actual positives that were detected.
Detailed explanation
Precision and recall are classification metrics based on a confusion matrix.
Recall and false positives are different aspects and can involve a trade-off.
Recall measures how much of the actual positive class was found.
Precision measures the correctness of positive predictions.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01ガイドのDomain 3.3で分類評価指標の意味を確認する。
Official references
What to confirm
適合率と再現率を分母の違いで説明し、見逃し・誤検知との関係を整理できる。
Key points
- Precision
- Recall
- Confusion matrix
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Threshold trade-off
Changing a decision threshold can improve one metric while worsening another.
Confusion matrix
A table that counts true and false positives and negatives for a classifier.