Using Random Forest to Learn Imbalanced Data

← Back to document page

Ling & Li (1998) over-sample the minority class by replicating the minority samples so that they attain the same size as the majority class. Over-sampling does not increase information; however by replication it raises the weight of the minority samples. Chawla et al. (2002) combine over-sampling and down-sampling

  Forest, Random, Attain, Random forests