Skip to content
BENIBIX Learn

Recall (Machine Learning)

AI📏 Evaluation & BenchmarksClassification Metrics

Also known as Recall, Sensitivity, True Positive Rate

Out of all the real yes cases, how many a model finds. Few misses means high recall.

Think of it like

It is like an egg hunt. Recall is how many of the hidden eggs you found.

Example

There are 50 fraud payments in Jack's data. His model catches 45 of them, so its recall is 90%, even though it also flags 30 honest ones.

Why it matters

It tells you how much slips through. That matters when a miss is worse than a false alarm.

Definition

Recall is how many of the real yes cases a model finds. Out of all the cases that truly were yes, it is the share the model caught. A cancer test with 90% recall finds 9 out of 10 people who have the illness. High recall means few missed cases. It matters most when a miss is costly, such as missing an illness or a fraud.
The filled dots are the real yes cases. Recall is the share the model's circle caught.

Common mistake

Pushing recall up and not watching precision. A model that says yes to all things has 100% recall and is of no use.

Key takeaways

  1. 1
    Share found. Of the real yes cases.
  2. 2
    Few misses. High recall.
  3. 3
    Pairs with precision. One up, one down.

Compare