Credit Card Fraud Detection Using Autoencoder Neural Network
This addresses the issue of imbalanced data classification, particularly for credit card fraud detection, but is incremental as it builds on existing denoising autoencoder and oversampling techniques.
The paper tackled the problem of noise in oversampling for imbalanced data classification by proposing a denoising autoencoder neural network (DAE) algorithm that oversamples minority classes and denoises the dataset, resulting in improved classification accuracy for minority classes compared to existing methods.
Imbalanced data classification problem has always been a popular topic in the field of machine learning research. In order to balance the samples between majority and minority class. Oversampling algorithm is used to synthesize new minority class samples, but it could bring in noise. Pointing to the noise problems, this paper proposed a denoising autoencoder neural network (DAE) algorithm which can not only oversample minority class sample through misclassification cost, but it can denoise and classify the sampled dataset. Through experiments, compared with the denoising autoencoder neural network (DAE) with oversampling process and traditional fully connected neural networks, the results showed the proposed algorithm improves the classification accuracy of minority class of imbalanced datasets.