Machine Learning Fundamentals · Model evaluation

Class imbalance handling

Techniques for when one class vastly outnumbers another, since accuracy alone becomes meaningless (predicting the majority class always "wins").

Also known as: oversampling, undersampling, SMOTE.

Start a lesson on Class imbalance handling

Learn first

• Classification evaluation metrics: Accuracy alone can lie; precision, recall, F1, and ROC-AUC each capture a different way a classifier can be right or wrong.

Related topics

• Ensemble methods: Combining many weak models (usually trees) into one stronger one, either by averaging (bagging) or correcting each other's mistakes in sequence (boosting).