Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • This Month
  • Published:

POINTS OF SIGNIFICANCE

The class imbalance problem

Nature Methods volume 18, pages 1270–1272 (2021)Cite this article

The exception proves the rule.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Rebalancing methods applied to a dataset with M = 50 normal and m = 5 diseased cells (IR = 10).
Fig. 2: Benefits of rebalancing classes in the training set by undersampling with increasing imbalance ratio on the performance of a CART decision tree classifier.
Fig. 3: Training with AUROC can provide acceptable performance at IR = 10 without rebalancing.
Fig. 4: Comparison of rebalancing methods on the performance of a CART decision tree classifier.

References

  1. Lever, J., Krzywinski, M. & Altman, N. Nat. Methods 13, 541–542 (2016).

    Article  CAS  Google Scholar 

  2. Krzywinski, M. & Altman, N. Nat. Methods 14, 757–758 (2017).

    Article  CAS  Google Scholar 

  3. Lever, J., Krzywinski, M. & Altman, N. Nat. Methods 13, 603–604 (2016).

    Article  CAS  Google Scholar 

  4. King, G. & Zeng, L. Polit. Anal. 9, 137–163 (2001).

    Article  Google Scholar 

  5. Altman, N. & Krzywinski, M. Nat. Methods 18, 840–842 (2021).

    Article  CAS  Google Scholar 

  6. Chawla, N. V., Bowyer, K. W., Hall, L. O. & Kegelmeyer, W. P. J. Artif. Intell. Res. 16, 321–357 (2002).

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Farmer School of Business, Miami University, Oxford, OH, USA

    Fadel M. Megahed

  2. Department of Mathematics, University of Dayton, Dayton, OH, USA

    Ying-Ju Chen

  3. IBM Research – Almaden, San Jose, CA, USA

    Aly Megahed & Yuya Ong

  4. Department of Statistics, The Pennsylvania State University, State College, PA, USA

    Naomi Altman

  5. Canada’s Michael Smith Genome Sciences Centre, Vancouver, British Columbia, Canada

    Martin Krzywinski

Authors
  1. Fadel M. Megahed
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ying-Ju Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Aly Megahed
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yuya Ong
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Naomi Altman
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Martin Krzywinski
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Martin Krzywinski.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Megahed, F.M., Chen, YJ., Megahed, A. et al. The class imbalance problem. Nat Methods 18, 1270–1272 (2021). https://doi.org/10.1038/s41592-021-01302-4

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41592-021-01302-4

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing