recall · shipping · also sensitivity, true positive rate, TPR

Recall

Recall is the fraction of observed positive outcomes identified by a classifier at a specified decision threshold.

Related Observatory article: Recall on hesela.com

For binary failure prediction, recall = TP / (TP + FN), where FN counts observed positives missed by the classifier. It answers how many labeled failures were detected, not how many alerts were useful. If the evaluation sample has no positive outcomes, recall is undefined; report the software convention. [1]

Relations

related

precision, false-positive-rate, class-prevalence, average-precision, smart

Limitations

• High recall alone does not bound false alarms: a classifier that labels every observation positive has recall one whenever positives exist. [1, 2]

Citations

1. scikit-learn: recall_score; binary recall and zero-division behavior. (accessed 2026-10-12)
2. scikit-learn: precision_recall_curve; the all-positive endpoint. (accessed 2026-10-12)
3. Hesela: disk failure prediction, metric denominators and a synthetic alert-count example. (accessed 2026-10-12)

JSON: /data/terms/recall.json