Toward Scalable Learning with Non-Uniform Class and Cost Distributions: A Case Study in Credit Card Fraud Detection

By Philip Ka-Fai Chan, Salvatore J. Stolfo. 1998

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

Data preprocessing techniques for classification without discrimination (2011, background): “In Cost-Sensitive and Utility-Based learning [5,12,20,28,31], it is assumed that not all types of prediction errors are equal and not all examples are as important.”
Three naive Bayes approaches for discrimination-free classification (2010, background): “In Cost-Sensitive and Utility-Bases learning [3, 5, 9], it is assumed that not all types of prediction errors are equal and not all examples are as important.”
A systematic study of the class imbalance problem in convolutional neural networks (2017, background): “ng set than other classes. This dierence is referred to as class imbalance. There are plenty of examples in domains like computer vision [6, 7, 8, 9, 10], medical diagnosis [11, 12], fraud detection [13] and others [14, 15, 16] where this issue is highly signicant and the frequency of one class (e.g”
The Influence of Class Imbalance on Cost-Sensitive Learning: An Empirical Study (2006, background): “As far as we know, there is only one work concerned the influence of class-imbalance on cost-sensitive learning [ 2 ].”
Deep learning detecting fraud in credit card transactions (2018, unknown): “used a number of supervised learning approaches in credit card fraud detection [18].”
Statistical fraud detection: A review (2002, background): “Issues such as these are discussed by Chan and Stolfo (1998) and”
Mining with rarity: a unifying framework (2004, background): “[9] P. K. Chan, and S. J. Stolfo.”
Does cost-sensitive learning beat sampling for classifying rare classes? (2005, methodology): “Some of these enhancements include introducing new “synthetic” examples when upsampling [5], deleting less useful majority-class examples when down-sampling [11] and using multiple sub-samples when downsampling such than each example is used in at least one subsample [3].”
Under Consideration for Publication in Knowledge and Information Systems Improving Svm Classification on Imbalanced Time Series Data Sets with Ghost Points (unknown): “There are many real-world applications where the data sets are highly imbalanced, such as oil spill detection from satellite images [16], credit card fraud detection [7], medical diagnostics [21], and predicting telecommunication equipment failure [31].”
On the Class Imbalance Problem (2008, background): “Chan and Stolfo [47] run a set of preliminary experiments to identify a good class distributions and then do resampling to generate multiple training sets with the desired class distributions.”
Editorial: special issue on learning from imbalanced data sets (2004, background): “cused) sampling is evolving under the multiple classifier systems or ensembles domain [8][12][11][46][51][17][31][59].”
Handling imbalanced datasets: A review (2006, background): “This approach can be used with any learning method and Chan and Stolfo evaluate it using four different learning algorithms .”
Learning When Training Data are Costly: The Effect of Class Distribution on Tree Induction (2003, methodology): “We consider this to be particularly significant because previous research on the effect of class distribution on learning has not employed this, or any other, adjustment (Catlett, 1991; Chan & Stolfo, 1998; Japkowicz & Stephen, 2002).”
Mining intrusion detection alarms for actionable knowledge (2002, methodology): “Data mining for fraud detection is investigated by Fawcett and Provost [15], and by Chan and Stolfo [10].”
Data mining with imbalanced class distributions: concepts and methods (2009, methodology): “In [26], several samples are generated from the training set.”