Answers are grounded in this paper and cite their sources.
Building a classification model on imbalanced datasets can be a challenging endeavor. Models built on data where examples of one class are greatly outnumbered by examples of the other class(es) tend to sacrifice accuracy with respect to the underrepr
No takes yet. Share an insight, caveat, or question.
Seiffert et al. (2009) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: