Recall

Recall - Another classification metric (sensitivity) defined as the fraction of actual positives that were correctly identified. In trading signals, recall measures “Of all true buy opportunities, how many did the model catch?” Recall = TP/(TP+FN). High recall means the model catches most real trades, though it might include more false alarms. Precision and recall together (often summarized by F1-score) provide a balanced view of model quality in imbalanced forex data.