Why Accuracy Lies on Imbalance
The rare-class trap.
The Comfortable Lie
Accuracy feels like the obvious score: how many predictions you got right. On balanced data it works fine, but imbalance quietly breaks it.
What Imbalance Means
A dataset is imbalanced when one class is far rarer than the others. Think fraud, disease, or churn: the interesting event is the rare one.
All lessons in this course
- Why Accuracy Lies on Imbalance
- Resampling: SMOTE and Undersampling
- Class Weights and Thresholds
- Pick Metrics for Rare Events