What is the recall formula in binary classification?

• A. TP / (TP + FP)
• B. TP / (TP + FN)
• C. (TP + TN) / (TP + FP + FN + TN)
• D. TN / (TN + FP)

Answer: (C)

Recall, also called sensitivity or the true positive rate, measures how well a model identifies actual positive cases. It is calculated as the number of true positives divided by the total number of actual positives: TP / (TP + FN). This uses all actual positives in the denominator (TP + FN) and asks what fraction of those positives the model correctly finds (TP). The other formulas describe different concepts: precision is TP / (TP + FP), which gauges how many predicted positives are correct; accuracy is (TP + TN) / (TP + FP + FN + TN), which reflects overall correctness across both classes; specificity is TN / (TN + FP), which measures how well negatives are identified.