KBA: kernel boundary alignment considering imbalanced data distribution

By Gang Wu, Edward Y. Chang. IEEE Transactions on Knowledge and Data Engineering, 2005

394 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/40b7d406c02807339401d3f216f7744603eea15d

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. Francisco Herrera
2. Alberto Fernández
3. Victoria López
4. Vasile Palade
5. Bartosz Krawczyk
6. Mikel Galar
7. Edurne Barrenechea Tartas
8. Rukshan Batuwita

In Computer Vision and Pattern Recognition

1. Alberto Fernández
2. Nitesh V. Chawla
3. Vasile Palade
4. Edurne Barrenechea Tartas
5. Francisco Herrera
6. Sven Krasser
7. Humberto Bustince
8. Mikel Galar

Citations in the corpus

• A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches (2012, background): “The algorithm level (internal) approaches create or modify the algorithms that exist, to take into account the significance of positive examples [17]–[19].”
• Analysis of preprocessing vs. cost-sensitive learning for imbalanced classification. Open problems on intrinsic data characteristics (2012, background): “…ones to take the class-imbalance problem into consideration (Barandela, Sánchez, García, & Rangel, 2003; Ducange, Lazzerini, & Marcelloni, 2010; Wu & Chang, 2005; Xu, Chow, & Taylor, 2007) and external approaches that preprocess the data in order to diminish the effect of their class imbalance…”
• AdaBoost with SVM-based component classifiers (2008, neutral): “Several researchers (Kubat and Matwin, 1997; Wu and Chang, 2005; Akbani et al., 2004) have used the g-means”
• Learning from Imbalanced Data (2009, methodology): “Additionally, in [86] and [87], the kernel-boundary alignment (KBA) algorithm was proposed which is based on the idea of modifying the kernel matrix generated by a kernel function according to the imbalanced data distribution.”
• SVMs Modeling for Highly Imbalanced Classification (2009, background): “However, for a highly imbalanced classification, the majority class pushes the ideal decision boundary toward the minority class [6], [7].”
• Imbalanced Deep Learning by Minority Class Incremental Rectification (2018, neutral): “(2) Algorithmlevel: Modifying existing algorithms to give more emphasis on the minority classes [8,9,10,43,44,45,46,47,48,49,50].”
• ADASYN: Adaptive synthetic sampling approach for imbalanced learning (2008, methodology): “In [27], a kernel-boundaryalignment (KBA) algorithm based on the idea of modifying the kernel matrix according to the imbalanced data distribution was proposed to solve this problem.”
• FSVM-CIL: Fuzzy Support Vector Machines for Class Imbalance Learning (2010, methodology): “In addition to these methods, several kernel-modification methods for training SVMs with imbalanced datasets have been proposed in [6] and [7].”
• Borderline over-sampling for imbalanced data classification (2009, methodology): “For the two problems described above, the first one can be solved by approximating the ideal boundary with an interpolation technique as proposed in [20], while the second one can be overcome by using different parameters C [17], say C and C−, for the minority and majority classes in which C is sele”
• EUSBoost: Enhancing ensembles for highly imbalanced data-sets by evolutionary undersampling (2013, background): “The fi rst group addresses the modi fi cation of existing algorithms [55].”