ML internals: Synthetic Minority Oversampling (SMOTE) Technique
Domino Data Lab
MAY 20, 2021
This renders measures like classification accuracy meaningless. propose a different strategy where the minority class is over-sampled by generating synthetic examples. The use of multiple measurements in taxonomic problems. The class imbalance problem: Significance and strategies. In their 2002 paper Chawla et al.
Let's personalize your content