# Stop Oversampling for Class Imbalance Learning: A Critical Review

arXiv: https://arxiv.org/abs/2202.03579 · HTML: https://mlsanity.com/paper/2202.03579/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 190 references that are in the map.
- First posted on arXiv 4 Feb 2022.

## Builds on (most cited first, 13 of 13)

- [Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning](https://mlsanity.com/paper/1609.06570/) (2016)
- [Conditional Wasserstein GAN-based Oversampling of Tabular Data for Imbalanced Learning](https://mlsanity.com/paper/2008.09202/) (2020)
- [Distance and Similarity Measures Effect on the Performance of K-Nearest Neighbor Classifier -- A Review](https://mlsanity.com/paper/1708.04321/) (2017)
- [LoRAS: An oversampling approach for imbalanced datasets](https://mlsanity.com/paper/1908.08346/) (2019)
- [Dimensionality Invariant Similarity Measure](https://mlsanity.com/paper/1409.0923/) (2014)
- [WOTBoost: Weighted Oversampling Technique in Boosting for imbalanced learning](https://mlsanity.com/paper/1910.07892/) (2019)
- [Visual Speech Recognition](https://mlsanity.com/paper/1409.1411/) (2014)
- [Deep Face Image Retrieval: a Comparative Study with Dictionary Learning](https://mlsanity.com/paper/1812.05490/) (2018)
- [Detailed Investigation of Deep Features with Sparse Representation and Dimensionality Reduction in CBIR: A Comparative Study](https://mlsanity.com/paper/1811.09681/) (2018)
- [Rule-and Dictionary-based Solution for Variations in Written Arabic Names in Social Networks, Big Data, Accounting Systems and Large Databases](https://mlsanity.com/paper/1502.05441/) (2015)
- [A Novel Hybrid Method for Network Anomaly Detection Based on Traffic Prediction and Change Point Detection](https://mlsanity.com/paper/1801.05309/) (2018)
- [Operator and Manufacturer Independent D2D Private Link for Future 5G Networks](https://mlsanity.com/paper/1911.00303/) (2019)
- [Intelligent Methods for Accurately Detecting Phishing Websites](https://mlsanity.com/paper/2002.07223/) (2020)

## Built on by (most cited first, 4 of 4)

- [Balancing the Scales: A Comprehensive Study on Tackling Class Imbalance in Binary Classification](https://mlsanity.com/paper/2409.19751/) (2024)
- [Automated Classification of Cybercrime Complaints using Transformer-based Language Models for Hinglish Texts](https://mlsanity.com/paper/2412.16614/) (2024)
- [Towards High Supervised Learning Utility Training Data Generation: Data Pruning and Column Reordering](https://mlsanity.com/paper/2507.10088/) (2025)
- [Beyond Synthetic Augmentation: Group-Aware Threshold Calibration for Robust Balanced Accuracy in Imbalanced Learning](https://mlsanity.com/paper/2509.02592/) (2025)

## Abstract (arXiv)

For the last two decades, oversampling has been employed to overcome the challenge of learning from imbalanced datasets. Many approaches to solving this challenge have been offered in the literature. Oversampling, on the other hand, is a concern. That is, models trained on fictitious data may fail spectacularly when put to real-world problems. The fundamental difficulty with oversampling approaches is that, given a real-life population, the synthesized samples may not truly belong to the minority class. As a result, training a classifier on these samples while pretending they represent minority may result in incorrect predictions when the model is used in the real world. We analyzed a large number of oversampling methods in this paper and devised a new oversampling evaluation system based on hiding a number of majority examples and comparing them to those generated by the oversampling process. Based on our evaluation system, we ranked all these methods based on their incorrectly generated examples for comparison. Our experiments using more than 70 oversampling methods and three imbalanced real-world datasets reveal that all oversampling methods studied generate minority samples that are most likely to be majority. Given data and methods in hand, we argue that oversampling in its current forms and methodologies is unreliable for learning from class imbalanced data and should be avoided in real-world applications.