Precision and Recall Intuitive Mnemonic

Remember the precision and recall definitions with this visual story.
Precision and Recall Intuitive Mnemonic
Precision and Recall Intuitive Mnemonic

Precision and recall are fundamental metrics used in machine learning classification. However, it may be hard to remember the definitions. Below is a mnemonic that could help you remember or learn precision and recall better.

You pick good apples and put them into your basket, while throwing the bad ones into the trash bucket. However, you have a hard time quickly recognizing the good apples from the bad ones quickly. So, you often have to guess. Once you are done, you would like to measure how well you guessed.

There are many ways to measure, but people often use precision and recall metrics for classification problems like this.

Precision Mnemonic

Am I precise and selected truly good apples into the basket? Precisely, I don’t select any bad apple into the basket! Good apples is precisely what I selected? Of all the apples I picked (selected), how many are actually good?

The basket of selected (relevant, n_selected) apples are the positive predictions.

n_selected = n_true_positives + n_false_positives

So precision is:

precision = true_positives / n_selected
precision = true_positives / (true_positives + n_false_positives)

Precision is related to quality.

Recall Mnemonic

Did I have sensitive recall and selected all the good apples from the good apples? Recall, I don’t forget any good apple from the tree! I selected all the good apples with total recall? Of all the good apples available, how many did I actually pick?

The good apples are the positive (n_positives) samples.

n_positives = n_true_positives + n_false_negatives

So recall is:

recall = true_positives / n_positives
recall = true_positives / (true_positives + n_false_negatives)

Recall is the same as sensitivity and is related to completeness, quantity.

picking-apples, family, robot, rural

Created on 06 Jul 2024. Updated on: 06 Jul 2024.
Thank you

About Vaclav Kosar How many days left in this quarter? Twitter Bullet Points to Copy & Paste Averaging Stopwatch Privacy Policy
Copyright © Vaclav Kosar. All rights reserved. Not investment, financial, medical, or any other advice. No guarantee of information accuracy.