Extreme re-balancing for SVMs: a case study

By Bhavani Raskutti, Adam Kowalczyk. 2004

350 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.

In Artificial Intelligence

Citations in the corpus

A systematic study of the class imbalance problem in convolutional neural networks (2017, background): “Raskutti and Kowalczyk [78] has shown on imbalanced genomic data that OSVM outperforms two-class SVM, yet Yousef et al.”
A Systematic Review on Imbalanced Data Challenges in Machine Learning (2019, background): “In Raskutti and Kowalczyk (2004), the authors proposed imbalance data handling solutions with highly unequal distributions and suggested that one-class learning from class containing positive samples only, yields perfect performance dealing with highly imbalanced data.”
Outlier Analysis (2013, methodology): “SVM methods are particularly popular for one-class classification [250, 323, 382, 396, 445].”
Loss Max-Pooling for Semantic Image Segmentation (2017, methodology): “Such methods were recently investigated [7, 35, 44, 45] for deep learning, some of them following ideas previously applied in shallow learning methods like random forests [27, 28] or support vector machines [38, 42].”
A Survey of Predictive Modeling on Imbalanced Domains (2016, methodology): “Developments made in this context include one-class SVMs (e.g., Sch¨olkopf et al. [2001], Manevitz and Yousef [2002], Raskutti and Kowalczyk [2004], Zhuang and Dai [2006a, 2006b], and Lee and Cho [2006]) and the use of an autoencoder (or autoassociator) (e.g., Japkowicz et al. [1995] and Japkowicz…”
One-Class Collaborative Filtering (2008, background): “A comparison of the two strategies can be found in [20].”
Selecting Representative Data Sets (2012, background): “Usefulness of the one-class learning on extremely unbalanced data sets composed of high dimensional noisy features is showed in [76].”
Handling imbalanced datasets: A review (2006, background): “Imbalanced data sets exists in many real-world domains, such as spotting unreliable telecommunication customers, detection of oil spills in satellite radar images, learning word pronunciations, text classification, detection of fraudulent telephone calls, information retrieval and filtering tasks, a”
Mind the gaps: weighting the unknown in large-scale one-class collaborative filtering (2009, background): “A comparison of the two strategies can be found in [14].”
Borderline over-sampling for imbalanced data classification (2009, background): “Raskutti and Kowalczyk [16] conducted one-class SVMs to learn only from positive class instances.”
A study of the behaviour of linguistic fuzzy rule based classification systems in the framework of imbalanced data-sets (2008, methodology): “(2) Solutions at the algorithmic level: In this case we may adjust our method by modifying the cost per class [32], adjusting the probability estimation in the leaves of a decision tree (establishing a bias towards the positive class) [43], or learning from just one class [33] (“recognition based le”
Data mining with imbalanced class distributions: concepts and methods (2009, background): “[29] notice that, in their experiments, one class SVMs outperforms traditional SVM in imbalanced domains.”