📏 Recall

Recall#

The share of actual positives the model correctly identifies.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

Recall — also the true positive rate or sensitivity — is the share of actual positives the model correctly catches:

\[\text{Recall} = \frac{TP}{TP + FN}.\]

It answers “of everything that was truly positive, how much did we find?” and falls when false negatives pile up.

The trade-off#

Recall trades off against precision — loosening the threshold catches more positives (higher recall) but admits more false alarms (lower precision), which is why they’re read together via the F1-score. High recall matters most when a miss is costly: disease screening, fraud, safety.

Its fairness role#

Comparing recall across groups is exactly the equal-opportunity fairness test — it asks whether qualified members of every group have the same chance of being correctly selected. Unequal recall means the model misses true positives more often for one group, a common and consequential bias.

Theme: Classification & Averaging Metrics · All terminology

Hint

Mind map — connected ideas

Fairness parity · Selection Rate · Precision (a.k.a. Positive Predictive Value, PPV) · F1-score · Macro Recall · Micro Recall

Hint

More in Classification & Averaging Metrics

Accuracy · AUC (Area Under the Curve) · Average Precision (AP) · Binary Classification · Classification Probability · Discriminatory Power · F1-score · Gini Coefficient · Harmonic Mean · Log Loss (also called Logarithmic Loss or Cross-Entropy Loss) · Macro AUC · Macro AUROC (Macro-Averaged AUROC) · Macro Averaging · Macro F1