Overview
SMOTE is a sophisticated over-sampling technique. Instead of simply duplicating existing minority instances, it creates new ones by selecting a minority instance and its nearest neighbors, then creating new points along the lines connecting them.
Benefits
- Reduces the risk of overfitting compared to random over-sampling.
- Provides more diverse data for the model to learn from.
Use Cases
- Fraud detection.
- Rare disease diagnosis.
- Predicting equipment failure.