article thumbnail

ML internals: Synthetic Minority Oversampling (SMOTE) Technique

Domino Data Lab

In their 2002 paper Chawla et al. 2002) have performed a comprehensive evaluation of the impact of SMOTE- based up-sampling. 2002) provide an example that illustrates the modifications. Generation of artificial examples. propose a different strategy where the minority class is over-sampled by generating synthetic examples.