AI engineering glossary · Evaluation metrics
What is TPR (recall)?
Recall (also called sensitivity, true positive rate, TPR, hit rate in classification): The share of all real positive cases that the model manages to find: TP / (TP + FN). It answers 'of everything that was truly there, how much did we catch?'. Catching most cases matters when missing one is costly, as in disease screening or fraud detection.
Recall: an example
There are 20 fraudulent transactions and the model flags 15 of them, so recall = 15 / 20 = 0.75.
Related terms
More Evaluation metrics terms
- Recall at k
- Reciprocal rank
- Relevance judgments
- Reliability diagram
- Residual plot
- ROC curve
- Root mean squared error
- Root mean squared logarithmic error
Where you learn it
Recall is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.