AI & Machine Learning · Intermediate

Class Imbalance

A dataset where one class is far more frequent than others, which can bias models toward the majority class.

Definition

With imbalance, accuracy is misleading because predicting the majority class scores well. Remedies include resampling (oversampling the minority, undersampling the majority, SMOTE), class weights, threshold tuning and metrics such as precision, recall, F1 and PR AUC.

Related terms