What is Recall In Machine Learning?

Recall

Recall is the fraction of values predicted to be of a positive class out of all the values that truly belong to the positive class (including false negatives) recall = predicted true positives / (true positives + false negatives)

Example

There are 100 credit card transactions; 90 transactions are legitimate (positive class) and 10 transactions are fraudulent. If your model predicts that 80 transactions are legitimate and 20 transactions are fraudulent, its recall is:
88.89% = 80 true positives / (80 true positives + 10 false negatives)

Recall graphic

Sign up for our monthly newsletter, The Drift.

Subscribe