Glossary term
Glossary term
Foundations
The less common label in a class-imbalanced dataset. For example, given a dataset containing 99% negative labels and 1% positive labels, the positive labels are the minority class.
Contrast with majority class.
Click the icon for additional notes.
See Datasets: Imbalanced datasets in Machine Learning Crash Course for more information.
Created for this library
A fraud team designs training so the minority class of fraudulent transactions still drives a meaningful share of the gradient signal.
A medical screening team uses class weights so the minority class of rare disease cases influences training proportionally.
A retention team uses oversampling to give the minority class of churned customers more weight during training.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License