Skip to content
BENIBIX Learn

Precision vs Recall

People often think precision and recall are two words for how good a model is. They measure two different mistakes. Precision asks: when the model says yes, how often is it right? Recall asks: of all the real yes cases, how many did it find? Raising one usually lowers the other, so you choose based on which mistake costs more.

AI📏 Evaluation & Benchmarks

Precision (Machine Learning)

When a model says yes, how often it is right. Few false alarms means high precision.

Precision is how often a model is right when it says yes. Out of all the cases it marked as yes, it is the share that really were yes. A spam filter with 95% precision is wrong only 5 times in 100 when it calls something spam. High precision means few false alarms. It matters most when a false alarm is costly, such as sending a real email to the spam folder.

Example
Emma's spam filter flags 200 emails. 190 really are spam and 10 are real mail, so its precision is 95%.
Why it matters
It tells you how far you can trust a yes. That matters when acting on a wrong yes hurts someone.
Full entry
AI📏 Evaluation & Benchmarks

Recall (Machine Learning)

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

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.

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.
Full entry

In one line

Precision (Machine Learning)
When a model says yes, how often it is right. Few false alarms means high precision.
Recall (Machine Learning)
Out of all the real yes cases, how many a model finds. Few misses means high recall.