Class Balancing Methods for Fraud Detection using Deep Learning
Kanika Kanika, Jimmy Singla · 2022 Second International Conference on Artificial Intelligence and Smart Energy (ICAIS) · 2022
The handling of imbalanced datasets in real-world applications like fraud detection systems is a challenging task. In most cases, the number of legitimate transactions exceeds the fraudulent transactions. The class imbalance problem can be solved through various means, such as data-level methods and algorithmic-level approaches. This paper reviews the various techniques used to handle the imbalanced datasets. Performance evaluation of the algorithmic-level and data-level class balancing methods has been performed. It has been found that algorithmic methods can be more efficient at handling the data in their imbalanced form.