ML internals: Synthetic Minority Oversampling (SMOTE) Technique
Domino Data Lab
MAY 20, 2021
We present the inner workings of the SMOTE algorithm and show a simple “from scratch” implementation of SMOTE. Their tests are performed using C4.5-generated 2002) do not present a rigorous mathematical treatment for this modification, and the suggested median correction appears to be purely empirical-driven.
Let's personalize your content