A Survey on GAN Techniques for Data Augmentation to Address the Imbalanced Data Issues in Credit Card Fraud Detection

Abstract

Data augmentation is an important procedure in deep learning. GAN-based data augmentation can be utilized in many domains. For instance, in the credit card fraud domain, the imbalanced dataset problem is a major one as the number of credit card fraud cases is in the minority compared to legal payments. On the other hand, generative techniques are considered effective ways to rebalance the imbalanced class issue, as these techniques balance both minority and majority classes before the training. In a more recent period, Generative Adversarial Networks (GANs) are considered one of the most popular data generative techniques as they are used in big data settings. This research aims to present a survey on data augmentation using various GAN variants in the credit card fraud detection domain. In this survey, we offer a comprehensive summary of several peer-reviewed research papers on GAN synthetic generation techniques for fraud detection in the financial sector. In addition, this survey includes various solutions proposed by different researchers to balance imbalanced classes. In the end, this work concludes by pointing out the limitations of the most recent research articles and future research issues, and proposes solutions to address these problems.

Sentiment analysis and opinion mining on educational data: A survey

“…• Data augmentation such as data synthesis or generation methods to augment the available training data can help address data scarcity and imbalance challenges [190]. This can involve generating synthetic data to supplement the biased or adversarial data, and oversampling minority or underrepresented groups to address data imbalance [191].…”

GANs-Based Intracoronary Optical Coherence Tomography Image Augmentation for Improved Plaques Characterization Using Deep Neural Networks

“…Our loss function for the discriminator was the sum of false and real image loss. Our aim was to diminish the error of forecasting real images of the dataset and the generated fake images [34,35] Generator and discriminator loss functions are already reported in the literature [36,37] and are expressed in Equations ( 1) and (2).…”

“…Both networks work against each other to generate perfect aliases of the original images. Generator and discriminator loss functions are already reported in the liter [36,37] and are expressed in Equations ( 1) and ( 2).…”

“…Both the discriminator and the generator were proficient enough to foster data close to the original data. During this training phase, discriminator gradients were conceded by the generator.Generator and discriminator loss functions are already reported in the literature[36,37] and are expressed in Equations (1) and (2).…”

A new hierarchical algorithm based on CapsGAN for imbalanced image classification

“…The third strategy involves utilizing deep learning networks, such as Convolutional Neural Networks (CNNs) [20] and Capsule Network (CapsNet) [23], or combining them with previous strategies [23]. The last strategy is image data augmentation, which is commonly used in deep learning and computer vision to artificially increase the size and diversity of the training dataset [24]. It includes common techniques such as rotation, scaling, cropping [25,26], as well as a combination of resampling and deep learning-based methods [23,27].…”