https://github.com/artefactory/smote_strategies_study
Do we need rebalancing strategies? A theoretical and empirical study around SMOTE and its variants (Sakho, Malherbe and Scornet; 2024)
https://github.com/artefactory/smote_strategies_study
Last synced: over 1 year ago
JSON representation
Do we need rebalancing strategies? A theoretical and empirical study around SMOTE and its variants (Sakho, Malherbe and Scornet; 2024)
- Host: GitHub
- URL: https://github.com/artefactory/smote_strategies_study
- Owner: artefactory
- License: mit
- Created: 2024-02-23T14:44:15.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-04-14T15:40:53.000Z (over 1 year ago)
- Last Synced: 2025-04-14T15:45:21.797Z (over 1 year ago)
- Language: Python
- Homepage:
- Size: 337 KB
- Stars: 9
- Watchers: 0
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## Do we need rebalancing strategies? A theoretical and empirical study around SMOTE and its variants.
Repository for [Do we need rebalancing strategies? A theoretical and empirical study around SMOTE and its variants](https://arxiv.org/pdf/2402.03819.pdf) paper.
In praticular, you will find code to reproduce the paper experiments. **If you want to use our proposed mthods, please use the following updated repository** : https://github.com/artefactory/mgs-grf
## ⭐ Table of Contents
- [Getting Started](#getting-started)
- [Data sets](#data-sets)
- [Acknowledgements](#acknowledgements)
## ⭐ Getting Started
If you want to reproduce our paper experiments:
- the notebooks [here](notebooks/classif_experiments.ipynb) and [here](notebooks/distances_experiments.ipynb) reproduce the experiments
- thise [code](./validation) contains implementation the protocols used for the numerical experiments of our article.
In order to use our MGS strategy:
- this [notebook](notebooks/resampling_example.ipynb) illustrates how to use it
- the strategy is implemented [here](./oversampling_strategies/)
## ⭐ Data sets
The data sets of used for our article should be dowloaded inside the *data/externals* folder. The data sets are available at the followings adresses :
* [Pima](https://www.kaggle.com/datasets/uciml/pima-indians-diabetes-database)
* Phoneme : https://github.com/jbrownlee/Datasets/blob/master/phoneme.csv
* Abalone : https://archive.ics.uci.edu/dataset/1/abalone
* Wine : https://archive.ics.uci.edu/dataset/186/wine+quality
* Haberman : https://archive.ics.uci.edu/dataset/43/haberman+s+survival
* Yeast : https://archive.ics.uci.edu/dataset/110/yeast
* Vehicle : https://archive.ics.uci.edu/dataset/149/statlog+vehicle+silhouettes
* Ionosphere : https://archive.ics.uci.edu/dataset/52/ionosphere
* Breast cancer Wisconsin : https://archive.ics.uci.edu/dataset/15/breast+cancer+wisconsin+original
* CreditCard : https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud
* MagicTel : https://www.openml.org/d/44125
* California : https://www.openml.org/d/44090
* House_16H : https://openml.org/d/821
## ⭐ Acknowledgements
This work was done through a partenership between **Artefact Research Center** and the **Laboratoire de Probabilités Statistiques et Modélisation** (LPSM) of Sorbonne University.
If you find the code usefull, please consider citing us :
```bib
@article{sakho2024we,
title={Do we need rebalancing strategies? A theoretical and empirical study around SMOTE and its variants},
author={Sakho, Abdoulaye and Malherbe, Emmanuel and Scornet, Erwan},
journal={arXiv preprint arXiv:2402.03819},
year={2024}
}
```