Opposite approach of Oversampling, removing some examples of the majority class from the training set. ![[CleanShot 2025-01-14 at 11.38.48@2x.png]]
The **** can be done randomly; that is, the examples to remove from the majority class can be chosen at random.
Alternatively, examples to withdraw from the majority class can be selected based on some property. One such property is Tomek Links.
Underfitting