Recall

Recall, also called sensitivity or true-positive rate, measures how many real positive cases a classifier detects. It is calculated as true positives divided by all actual positives, including those the model missed. High recall is important when false negatives are especially harmful, such as failing to detect a disease, threat, or safety defect. Lowering the decision threshold can improve recall but usually reduces precision by generating more false positives. Like any metric, recall depends on representative labels and a clear definition of the positive class. Teams should examine precision–recall tradeoffs, subgroup results, operational capacity, and the consequences of both missed cases and unnecessary interventions.