CoChat

Optimal selection of resampling methods for imbalanced data with high complexity

Open access
Kim, Annie, Jung, Inkyung · 2023 · PLOS ONE

The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.

Abstract

Class imbalance is a major problem in classification, wherein the decision boundary is easily biased toward the majority class. A data-level solution (resampling) is one possible solution to this problem. However, several studies have shown that resampling methods can deteriorate the classification performance. This is because of the overgeneralization problem, which occurs when samples produced by the oversampling technique that should be represented in the minority class domain are introduced into the majority-class domain. This study shows that the overgeneralization problem is aggravated in complex data settings and introduces two alternate approaches to mitigate it. The first approach involves incorporating a filtering method into oversampling. The second approach is to apply undersampling. The main objective of this study is to provide guidance on selecting optimal resampling methods in imbalanced and complex datasets to improve classification performance. Simulation studies and real data analyses were performed to compare the resampling results in various scenarios with different complexities, imbalances, and sample sizes. In the case of noncomplex datasets, undersampling was found to be optimal. However, in the case of complex datasets, applying a filtering method to delete misallocated examples was optimal. In conclusion, this study can aid researchers in selecting the optimal method for resampling complex datasets.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0288540

Cites

33 of 33 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig 1 Visual representation of SMOTE. xi: Randomly selected minority class sample; xzi: Instance close to xi; xnew: New artificial example generated by interpolation between two instances.
Fig 1 Visual representation of SMOTE. xi: Randomly selected minority class sample; xzi: Instance close to xi; xnew: New artificial example generated by interpolation between two instances.CC BY · PubMed Central
Fig 2 Illustration of TL.
Fig 2 Illustration of TL.CC BY · PubMed Central
Fig 3 Backbone model under low complexity (c = 1). MIN: Minority class; MAJ: Majority class.
Fig 3 Backbone model under low complexity (c = 1). MIN: Minority class; MAJ: Majority class.CC BY · PubMed Central
Fig 4 Backbone model under medium complexity (c = 2).
Fig 4 Backbone model under medium complexity (c = 2).CC BY · PubMed Central
Fig 5 Backbone model under extreme complexity (c =2, but classes are spaced apart).
Fig 5 Backbone model under extreme complexity (c =2, but classes are spaced apart).CC BY · PubMed Central
Fig 6 Differences between ROC and PR curves.
Fig 6 Differences between ROC and PR curves.CC BY · PubMed Central
Fig 7 Differences in AUPRC ranks under low complexity. The star markers at the top of bars indicate significant performance gains.
Fig 7 Differences in AUPRC ranks under low complexity. The star markers at the top of bars indicate significant performance gains.CC BY · PubMed Central
Fig 8 Differences in AUPRC ranks under medium complexity. The star markers at the top of bars indicate significant performance gains.
Fig 8 Differences in AUPRC ranks under medium complexity. The star markers at the top of bars indicate significant performance gains.CC BY · PubMed Central
Fig 9 Difference in AUPRC ranks under extreme complexity. The star markers at the top of bars indicate significant performance gains.
Fig 9 Difference in AUPRC ranks under extreme complexity. The star markers at the top of bars indicate significant performance gains.CC BY · PubMed Central
Fig 10 Mean differences in AUPRC values using F3. The more complex the dataset, the darker the color of the bar.
Fig 10 Mean differences in AUPRC values using F3. The more complex the dataset, the darker the color of the bar.CC BY · PubMed Central
Fig 11 Mean differences in AUPRC values using N2. The more complex the dataset, the darker the color of the bar.
Fig 11 Mean differences in AUPRC values using N2. The more complex the dataset, the darker the color of the bar.CC BY · PubMed Central
Fig 12 Mean differences in AUPRC values using C2. The more complex the dataset, the darker the color of the bar.
Fig 12 Mean differences in AUPRC values using C2. The more complex the dataset, the darker the color of the bar.CC BY · PubMed Central
Fig 13 Complex and noncomplex areas in real datasets.
Fig 13 Complex and noncomplex areas in real datasets.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 33 references

Research works printed but not linked: 8

  1. He H, Ma Y, editors. Imbalanced learning: foundations, algorithms, and applications. pp. 187.
  2. Herawan T, Deris MM, Abawajy J, editors. Proceedings of the first international conference on advanced data and information engineering (DaEng-2013). Springer Science & Business Media; 2013 Dec 14.
  3. Chen XW, Gerlach B, Casasent D. Pruning support vectors for imbalanced data classification. InProceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005. 2005 Jul 31 (Vol. 3, pp. 1883– 1888). IEEE.
  4. Yan R, Liu Y, Jin R, Hauptmann A. On predicting rare classes with SVM ensembles in scene classification. In2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings.(ICASSP’03). 2003 Apr 6 (Vol. 3, pp. III– 21). IEEE.4
  5. He H, Bai Y, Garcia EA, Li S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence) 2008 Jun 1 (pp. 1322– 1328). IEEE.
  6. Tomek I., “ Two modifications of CNN,” In Systems, Man, and Cybernetics, IEEE Transactions on, vol. 6, pp 769– 772, 1976.
  7. Wilson D., Asymptotic” Properties of Nearest Neighbor Rules Using Edited Data,” In IEEE Transactions on Systems, Man, and Cybernetrics, vol. 2 (3), pp. 408– 421, 1972.
  8. Davis J, Goadrich M. The relationship between Precision-Recall and ROC curves. InProceedings of the 23rd international conference on Machine learning 2006 Jun 25 (pp. 233– 240).

Cited by

Metadata

Authors
Kim, A., & Jung, I.
Affiliations
Yonsei University (KR)
Corresponding author
Jung, Inkyung
Venue
PLOS ONE
Published
2023-07-27
Volume
18
Issue
7
Pages
e0288540
ISSN
1932-6203 (ISSN-L)
OpenAlex
W4385322723
Cited by
28
References
31
Citations by year
2026: 9 · 2025: 13 · 2024: 6
FWCI
2.64
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • YUHSpace (Yonsei University Medical Library) (submittedVersion, cc-by-nc-nd)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Imbalanced Data Classification Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
  • Social Sciences › Business, Management and Accounting › Marketing › Customer churn and segmentation
Keywords
class imbalance, resampling, oversampling, undersampling, overgeneralization, complex data, minority class, classification performance
MeSH terms
Algorithms, Computer Simulation, Sample Size

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.