Skip to content
Sign in

What is Recall?

Recall is an evaluation metric that measures the proportion of actual positive cases a model correctly identifies. It shows how well the model finds all relevant instances in the data.

Recall is calculated as true positives divided by the sum of true positives and false negatives. It focuses on minimizing missed positives rather than avoiding false alarms.

In binary classification it equals the true positive rate or sensitivity. High recall means few actual positives are overlooked, which matters when missing a case carries high cost.

Recall often trades off against precision; improving one can lower the other, so they are commonly balanced using the F1 score.

Example

In a medical test for cancer, recall tells what fraction of patients who actually have cancer were correctly flagged by the model. A recall of 0.95 means 95 percent of real cancer cases were detected.

Why it matters

Recall is critical in high-stakes domains such as disease screening and fraud detection where failing to catch a positive instance can have serious consequences.

Frequently asked questions

Recall measures how many actual positives were found; precision measures how many predicted positives were actually correct.