Class-Boundary Alignment for Imbalanced Dataset Learning

By Gang Wu, Edward Y. Chang. 2003

410 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Box drawings for learning with imbalanced data (2014, background): “…directly related to the goal of interpretable imbalanced classiﬁcation, speciﬁcally kernel and active learning methods (Raskutti and Kowalczyk 2004; Wu and Chang 2003), and work on sampling (Abe 2003; Chawla et al. 2002) that includes un-dersampling, oversampling, and data generation, which can be…”
Outlier Analysis (2013, background): “Numerous cost-sensitive variations of different classifiers have been proposed along the lines of weighting, and include the Bayes classifier [496], nearest neighbor classifier [506], decision trees [450, 463], rule-based classifiers [245, 247] and SVM classifiers [443, 470].”
A Survey of Predictive Modeling on Imbalanced Domains (2016, background): “Wu and Chang [2003] accomplish this with an algorithm that changes the kernel function.”
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].”
Applying Support Vector Machines to Imbalanced Datasets (2004, methodology): “For instance, it assumes that the space between two positive instances is assumed to be positive and the neighborhood of a positive instance is also assumed to be positive [15], which may not always be true.”
Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning (2005, methodology): “proposed classboundary alignment algorithm which modify the class boundary through changing the kernel function of SVMs [20].”
Modelling input texts: from Tree Kernels to Deep Learning (2015, background): “This can be explained by the fact, as suggested in [179], that ziC imposes only an upper bound on dual variables αi, which results in poorer flexibility to control the class-imbalance with the j parameter of SVM.”
Under Consideration for Publication in Knowledge and Information Systems Improving Svm Classification on Imbalanced Time Series Data Sets with Ghost Points (unknown): “This is because the class-boundary learned by the SVM is skewed toward the majority/negative class [ 34 ].”
Deep Learning Based Vulnerability Detection: Are We There Yet? (2020, background): “Lastly, models suffer from data imbalance [27] between vulnerable and benign code as the proportion of vulnerable examples in comparison to the neutral ones in real world”
On the Class Imbalance Problem (2008, methodology): “In [29] an algorithm is proposed by changing the kernel function to develop this bias.”
Learning from Imbalanced Data (2009, methodology): “For instance, [75] and [76] are good examples of”
Cost-Sensitive Learning of Deep Feature Representations From Imbalanced Data (2015, background): “However, they do not address the class imbalance learning of CNNs, which have recently emerged as the most popular tool for supervised classification, recognition and segmentation problems in computer vision [12, 22, 25, 26, 27].”
Adaptive Machine Learning for Credit Card Fraud Detection (2015, background): “Other studies have also reported the negative effect of skewed class distributions not only on decision trees [21, 82], but also for Neural Networks [82, 104], k-Nearest Neighbor (kNN) [100, 105, 106] and SVM [107, 108].”
Cost-sensitive boosting for classification of imbalanced data (2007, methodology): “For SVMs, proposals such as using different penalty constants for different classes [38], or adjusting the class boundary based on kernel-alignment ideal [13], are reported.”
Handling imbalanced datasets: A review (2006, background): “In [32] an algorithm is proposed that changes the kernel function to develop this bias.”