I find myself often forgetting precision and recall, and which one is which. So I am writing this blog post to remind myself, and hopefully help other people too.

Let's start with some Venn diagrams that illustrate the target labels and our predicted labels.

• TP = True Positive
• FP = False Positive
• TN = True Negative
• FN = False Negative

And the following figures will illustrate how those sections of the Venn diagram relate to accuracy, precision, and recall.

• Accuracy is used when the classes are balanced.

• Precision is used when the thing you are interested in identifying is very common.

• Recall is used when the thing you are interested in identifying is very rare.