Artificial Intelligence & Machine Learning
Recall
Definition
In classification tasks, recall (also known as sensitivity or true positive rate) is a metric that measures the proportion of actual positives that were correctly identified by the model. It answers the question: "Of all the actual positive cases, how many did the model find?"
Why It Matters
Recall is important when the cost of a false negative is high. You want to make sure you find all the positive cases, even if it means you get some false positives.
Contextual Example
In medical screening for a disease, you want high recall. It is better to have some healthy people be incorrectly flagged for more testing (false positives) than to miss someone who actually has the disease (a false negative).
Common Misunderstandings
- Recall = True Positives / (True Positives + False Negatives).
- There is often a trade-off between precision and recall.