Safe-Level-SMOTE: Safe-Level-Synthetic Minority Over-Sampling TEchnique for Handling the Class Imbalanced Problem

By Chumphol Bunkhumpornpat, Krung Sinapiromsaran, Chidchanok Lursinsap. Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2009

967 citations recorded by the source corpus.

Subfields: Accounting, Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/b012256861a72e6d046f454f8d867647c51ced0e

How researchers see this paper

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

In Accounting

In Artificial Intelligence

Citations in the corpus

An empirical survey of data augmentation for time series classification with neural networks (2020, methodology): “There have also been a number of improvements on SMOTE, such as Safe-Level-SMOTE [64], SMOTE based on the furthest neighbor (SMOTEFUNA) [65], Cost Minimization Oriented SMOTE (CMO-SMOTE) [66], Density-Based SMOTE (DBSMOTE) [67], etc.”
SMOTE for high-dimensional class-imbalanced data (2013, methodology): “Under the null hypothesis all the theoretical results presented in this paper would apply also for Borderline-SMOTE [22] and Safe-Level-SMOTE [23].”
Oversampling for Imbalanced Learning Based on K-Means and SMOTE (2017, methodology): “, 2001), safe areas (Bunkhumpornpat et al., 2009), or to areas close to the decision boundary (Han et al.”
Cross-Validation for Imbalanced Datasets: Avoiding Overoptimistic and Overfitting Approaches [Research Frontier] (2018, background): “[18], called Safe-Level-SMOTE, only synthesizes minority examples around safe regions.”
An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics (2013, methodology): “Other approaches such as Safe-Level-SMOTE [21] and ADASYN [65] work in a similar way.”
Loss Max-Pooling for Semantic Image Segmentation (2017, result): “In our experiments on the Cityscapes and Pascal VOC 2012 segmentation datasets we find consistently improved results, demonstrating the efficacy of our approach.”
DBSMOTE: Density-Based Synthetic Minority Over-sampling TEchnique (2012, background): “Bunkhumpornpat, Sinapiromsaran, and Lursinsap (2009) designed Safe-Level-SMOTE [6], which generates synthetic instances along the same line segment as SMOTE but locates them closer to a minority class than a majority class.”
GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks (2021, background): “Safe-Level-SMOTE [7] computes the safe level for each interpolation direction using majority class neighbors, in order to make the generated new samples safer.”
SMOTE-IPF: Addressing the noisy and borderline examples problem in imbalanced classification by a re-sampling method with filtering (2015, methodology): “Within this group, the Safe-Levels-SMOTE (SL-SMOTE) [8], the Borderline-SMOTE (B1-SMOTE and B2-SMOTE) [23] or LN-SMOTE [37] methods are found, which try to create positive examples close to areas with a high concentration of positive examples or only inside the boundaries of the positive class.”
On the use of MapReduce for imbalanced big data using Random Forest (2014, methodology): “Oversampling methods [11,12,26] aim to balance the class distribution adding to the new dataset instances from the minority class; undersampling methods [22,30,52] try to adjust the class proportion deleting instances from the majority class; and hybrid methods [6] combine the two previous approache”
SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary (2018, background): “The interpolation mechanisms can be range restricted (Han et al., 2005; Bunkhumpornpat et al., 2009; Maciejewski & Stefanowski, 2011), for example by looking not only for nearest neighbours from the minority class but also from majority class; creating new examples closer to the selected instance th”