Ensemble of deep sequential models for credit card fraud detection

In the recent years, the fast development of e-commerce technologies made it possible for people to select the most desirable items in terms of suggested price, quality and quantity among various services, facilities, shops and stores from all around the world. However, it also made it easier for fraudsters to abuse this huge opportunity. As credit card has become the most popular mode of payment, the fraudulent activities using credit card payment technologies are rapidly increasing as a result. Therefore, it is obligatory for financial institution to think of an automatic deterrent mechanism to prevent these fraudulent actions. Although many works have been done in this area using traditional statistical and machine learning methods, most of them do not take the sequential nature of transactional data into account. In this paper, we proposed an ensemble model based on sequential modeling of data using deep recurrent neural networks and a novel voting mechanism based on artificial neural network to detect fraudulent actions. In addition, we present a novel algorithm for training the aforementioned voting approach. Our experimental results on two real world datasets demonstrate that the proposed model outperforms the state-of-the-art models in all evaluation criteria. Moreover, the time analysis of the proposed model shows that it is more efficient in terms of real-time performance versus the recent models in the field. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

The rapid development of e-commerce technologies in recent years has provided more choices for people, but also increased the risk of fraud. This paper proposes an ensemble model based on deep learning and artificial neural networks to detect fraudulent actions, showing its advantage in real-time performance.