Skip to content
Sign in

What is Precision?

Precision is an evaluation metric for classification models that measures the proportion of true positive predictions among all positive predictions made.

It is calculated as true positives divided by the sum of true positives and false positives. This focuses only on the cases the model labeled as positive.

A high precision score means the model makes few false positive errors when predicting the positive class. It does not consider how many actual positives were missed.

Precision is commonly paired with recall because improving one often reduces the other; the balance between them is summarized by the F1 score.

Example

In a spam filter, if the model flags 100 emails as spam and 90 of them are actually spam, precision is 0.90. This shows that 90% of its spam predictions were correct.

Why it matters

Precision is critical in applications where false alarms are costly or disruptive, such as fraud detection or medical screening, helping practitioners control the reliability of positive alerts.

Frequently asked questions

Precision measures correctness among predicted positives while recall measures how many actual positives were found.