SMOTE-RSB*: a hybrid preprocessing approach based on oversampling and undersampling for high imbalanced data-sets using SMOTE and rough sets theory

By Enislay Ramentol, Yailé Caballero Mota, Rafael Bello, Francisco Herrera. Knowledge and Information Systems, 2012

453 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/9845e3f144376c349b6204e47fe425d10f672e91

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Octavio Loyola-González
2. Milton García-Borroto
3. J. A. Carrasco-Ochoa
4. Paula Branco
5. Rita P. Ribeiro
6. Luís Torgo
7. José Francisco Martínez Trinidad
8. Ferdous Sohel

In Computational Theory and Mathematics

1. Octavio Loyola-González
2. Milton García-Borroto
3. J. A. Carrasco-Ochoa
4. Mengchu Zhou
5. Paula Branco
6. Mohammed Bennamoun
7. José Francisco Martínez Trinidad
8. Rita P. Ribeiro

In Electrical and Electronic Engineering

1. Octavio Loyola-González
2. Haoyue Liu
3. Milton García-Borroto
4. J. A. Carrasco-Ochoa
5. Mengchu Zhou
6. Paula Branco
7. Munawar Hayat
8. Mohammed Bennamoun

Citations in the corpus

• Boosting methods for multi-class imbalanced data classification: an experimental review (2020, background): “Over-sampling approaches create copies of existing samples or add more samples to the minority class [27].”
• Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data (2015, background): “, oversampling (SMOTE [10]), undersampling (random undersampling [62]), and hybrid sampling (SMOTE-RSB∗ [13]).”
• SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary (2018, unknown): “, 2012) DBSMOTE 3 Graph based (Farquad & Bose, 2012) SVM-Balance 3 (Puntumapon & Waiyamai, 2012) TRIM-SMOTE 3 3 (Ramentol et al., 2012) SMOTE-RSB* 3 (Wang et al.”
• Study of the impact of resampling methods for contrast pattern based classifiers in imbalanced databases (2016, methodology): “SMOTE-RSB: Hybrid Preprocessing using SMOTE and Rough Sets Theory [61] is a hybrid method for preprocessing imbalanced datasets through the creation of new synthetic objects of the minority class using SMOTE and improving the quality of these new objects through editing techniques based on the Rough”
• A Survey of Predictive Modeling on Imbalanced Domains (2016, methodology): “…approach include the following: SMOTE and Tomek Links (SMOTE+Tomek) [Batista et al. 2004], SMOTE and Edited Nearest Neighbors (SMOTE+ENN) [Batista et al. 2004], SMOTE and Fuzzy Rough Set Theory (SMOTE+FRST) [Ramentol et al. 2012b] or SMOTE and Rough Sets Theory (SMOTE+RSB) [Ramentol et al. 2012a].”
• An embedded feature selection method for imbalanced data classification (2019, methodology): “Two standard methods have been used to deal with a class imbalance problem [2]: A) Data Level: Modify majority or minority class distribution to allow the classification algorithm to equally treat an imbalanced dataset.”