This document discusses addressing imbalanced data sets where the distribution of classes is unequal. Imbalanced data can cause models to be biased towards the majority class and perform poorly on the minority class. Various techniques are presented to address this issue including sampling methods like oversampling and undersampling, cost-sensitive learning, and ensemble methods. Two case studies on credit card fraud detection and medical diagnosis with imbalanced data sets are also summarized where combinations of these techniques improved model performance.