ADASYN: Adaptive synthetic sampling approach for imbalanced learning

He, Haibo; Bai, Yang; Garcia, E.A.; Li, Shutao

doi:10.1109/ijcnn.2008.4633969

Cited by 1,570 publications

(742 citation statements)

References 15 publications

Supporting

Mentioning

726

Contrasting

Unclassified

Order By: Relevance

“…Adaptive Synthetic Over-Sampling (ADASYN) is considered as an extension of SMOTE, which is characterized by the creation of more samples in the vicinity of the boundary in the midst of the two classes than inside the minority class [62].…”

Section: Over-sampling Methodsmentioning

confidence: 99%

“…Thus, more "intelligent" or heuristic sampling methods have been developed; for example, SMOTE [61] and ADASYN [62]. Moreover, they have been combined with editing methods, such as Tomek's Links (TL) [26], Editing Nearest Neighbor (ENN) [27], Condensed Nearest Neighbor rule (CNN) [28] and others [22,[29][30][31][32][33]63]; i.e., hybrid methods, such as SMOTE+TL, SMOTE+CNN, SMOTE+OSS and SMOTE+ENN [22,64,65].…”

Section: Sampling Class Imbalance Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Data Sampling Methods to Deal With the Big Data Multi-Class Imbalance Problem

et al. 2020

View full text Add to dashboard Cite

The class imbalance problem has been a hot topic in the machine learning community in recent years. Nowadays, in the time of big data and deep learning, this problem remains in force. Much work has been performed to deal to the class imbalance problem, the random sampling methods (over and under sampling) being the most widely employed approaches. Moreover, sophisticated sampling methods have been developed, including the Synthetic Minority Over-sampling Technique (SMOTE), and also they have been combined with cleaning techniques such as Editing Nearest Neighbor or Tomek's Links (SMOTE+ENN and SMOTE+TL, respectively). In the big data context, it is noticeable that the class imbalance problem has been addressed by adaptation of traditional techniques, relatively ignoring intelligent approaches. Thus, the capabilities and possibilities of heuristic sampling methods on deep learning neural networks in big data domain are analyzed in this work, and the cleaning strategies are particularly analyzed. This study is developed on big data, multi-class imbalanced datasets obtained from hyper-spectral remote sensing images. The effectiveness of a hybrid approach on these datasets is analyzed, in which the dataset is cleaned by SMOTE followed by the training of an Artificial Neural Network (ANN) with those data, while the neural network output noise is processed with ENN to eliminate output noise; after that, the ANN is trained again with the resultant dataset. Obtained results suggest that best classification outcome is achieved when the cleaning strategies are applied on an ANN output instead of input feature space only. Consequently, the need to consider the classifier's nature when the classical class imbalance approaches are adapted in deep learning and big data scenarios is clear.

show abstract

Section: Over-sampling Methodsmentioning

confidence: 99%

Section: Sampling Class Imbalance Approachesmentioning

confidence: 99%

Data Sampling Methods to Deal With the Big Data Multi-Class Imbalance Problem

et al. 2020

View full text Add to dashboard Cite

show abstract

“…SMOTE encounters some drawbacks including over-generalization and lack of systematizing disjuncts. Enhanced techniques such as Borderline-SMOTE [20], SafeLevel-SMOTE [21] and adaptive synthetic sampling (ADASYN) [22] help to overcome these drawbacks. The proposed technique follows the same baseline while leveraging the disjuncts and generalization issue.…”

Section: Related Workmentioning

confidence: 99%

Improved classification of large imbalanced data sets using rationalized technique: Updated Class Purity Maximization Over_Sampling Technique (UCPMOT)

Patil¹,

Sonavane²

2017

J Big Data

View full text Add to dashboard Cite

“…Due to imbalance of data in the training set, Adaptive Synthesis(ADASYN) [7] is applied to generate synthetic data for training. Parameters of ADASYN are: beta=1, k=5, which means after ADASYN, we will have the same amount of data labelled as 1 and -1.…”

Section: Dealing With Imbalanced Datamentioning

confidence: 99%