ADASYN: Adaptive synthetic sampling approach for imbalanced learning

By Haibo He, Yang Bai, Edwardo A. Garcia, Shutao Li. IEEE World Congress on Computational Intelligence, 2008

4,901 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/48234756b7cf798bfeb47328f7c5d597fd4838c2

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

In Statistics and Probability

Citations in the corpus

A review of ensemble learning and data augmentation models for class imbalanced problems: combination, implementation and evaluation (2023, background): “Adaptive synthetic sampling (ADASYN)[232] adaptively generates minority data samples according to their distributions.”
Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection (2021, background): “SMOTE [5] is a representative interpolation-based over-sampling methods and also has lots of extensions [12, 13].”
SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary (2018, background): “• ADASYN (He et al., 2008): Its main idea proceeds from the assumption of utilizing a weighted distribution depending on the type of minority examples according to their complexity for learning.”
DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data (2021, background): “While not many solutions have been proposed for guided undersampling [14], [15], [16], oversampling has gained much more attention due to the success of SMOTE [8], which led to the introduction of a plethora of variants [17], [18], [19].”
Oversampling Techniques for Bankruptcy Prediction: Novel Features from a Transaction Dataset (2018, methodology): “To overcome this limitation, ADASYN, which uses a systematic method to adaptively create different amounts of synthetic data based on their distributions, was proposed.”
Learning to Segment the Tail (2020, background): “The former aims to re-balance the training samples across classes [16, 3, 11, 6]; while the latter focuses on assigning different weights to adjust the loss function [18, 40, 47, 7].”
Striking the Right Balance With Uncertainty (2019, methodology): “, SMOTE NN [7], ADASYS NN [18]) and algorithm-level imbalance removal techniques (e.”
Conditional Wasserstein GAN-based Oversampling of Tabular Data for Imbalanced Learning (2020, background): “Secondly, they require no binary decision-threshold unlike metrics such as accuracy, balanced accuracy, or the f 1 -score, which is the harmonic mean between precision and recall.”
Imbalanced-learn: A Python Toolbox to Tackle the Curse of Imbalanced Datasets in Machine Learning (2016, methodology): “Method Over-sampling Under-sampling Binary Mutli-class Binary Multiclass ADASYN (He et al., 2008) 3 7 7 7 SMOTE (Chawla et al.”
A theoretical distribution analysis of synthetic minority oversampling technique (SMOTE) for imbalanced learning (2023, neutral): “…of SMOTE extensions have been developed such as: Borderline SMOTE (Han et al., 2005), Safe-level SMOTE (Bunkhumpornpat et al., 2009), ADASYN (He et al., 2008), SVM SMOTE (Nguyen et al., 2011), Localized Random Affine Shadowsampling (LoRAS) (Bej et al., 2021), CDSMOTE (Elyan et al., 2021),…”
CatBoost for big data: an interdisciplinary review (2020, methodology): “The algorithms used for addressing class imbalance in the data are Synthetic Minority Over-sampling Technique (SMOTE) [15] and Adaptive Synthetic Sampling Approach (ADASYN) [41].”
A Systematic Review on Imbalanced Data Challenges in Machine Learning (2019, methodology): “It proves to be an efficient way of handling imbalanced data in the following ways: Hybrid sampling (He et al. 2008), (Kubat et al. 1997), (Qian et al. 2014), (Charte et al. 2015) combines sampling methods reduces limitations of sampling methods (1) It tends to reduce the situation of imbalance…”
Learning a Unified Classifier Incrementally via Rebalancing (2019, background): “Previous efforts to tackle the class imbalance can be roughly divided into two groups: data resampling [13, 6, 12] and cost-sensitive learning [32, 17, 38, 16, 9].”
On the importance of normative data in speech-based assessment (2017, methodology): “We apply ADASYN on the training set only.”
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation (2021, background): “sampling [23,25,81], loss re-weighting [42,59], and transfer learning [36, 43].”
Solving Long-tailed Recognition with Deep Realistic Taxonomic Classifier (2020, background): “Sample synthesis [29, 65] has also been proposed to increase the population of tail classes.”