AI Glossary Term
What does “Precision” mean?
Definition
The proportion of predicted positives that are actually correct.
How Precision works
Example
A spam filter flags 100 emails and 90 really are spam. Its precision is 90 percent: of everything it flagged, nine in ten were correct. The ten wrongly flagged emails are the cost.
Why it matters
High precision means few false alarms. It matters most when a wrong flag is costly, such as blocking legitimate messages or accusing someone wrongly.
Common questions
Precision vs recall?
Precision asks how many flagged items were right. Recall asks how many of the real cases were found.
How is precision calculated?
True positives divided by the sum of true positives and false positives.
Can precision be gamed?
Yes. A model that flags only one very obvious case can have perfect precision but miss almost everything, so it is reported alongside recall.