A Novel Method for Detection of Fraudulent Bank Transactions using Multi-Layer Neural Networks with Adaptive Learning Rate

Document Type : Research Paper


Department of Electrical Engineering, Mahdishahr Branch, Islamic Azad University, Mahdishahr, Iran.


Fraud refers to earning wealth including property, goods and services through immoral and non-legal channels. Fraud has always been in action and has experienced an increasing trend worldwide. Fraud in financial transactions not only leads to the loss of huge financial resources but also leads to a reduction in the trust of customers in using modern banking systems and hence, a reduction in the efficiency of the systems and optimal management of financial transactions. In recent years, new means of fraud have been discovered by emerging new technologies in the banking industry. Although a new information system carries advantages and benefits, new opportunities are made for fraudsters. The applications of fraud detection methods encompass fraud detection in an organization, analysis of frauds and user/customer behaviour analytics to predict future behaviour and reduce fraud risks. In recent decades, employing new technologies in the management of banking transactions has risen. Banks and financial institutions inevitably migrated from traditional banking to modern online banking to provide effective services. Although the use of online banking systems improves the management of financial processes and speeds up services to customers of institutions, some issues can also be carried out. Financial fraud is one issue organizations seek to prevent and reduce its effects. In this paper, a novel machine learning-based model is presented to detect fraud in electronic banking transactions using profile data of bank customers. In the proposed method, transactional data from banks are leveraged and a multi-layer perceptron neural network with an adaptive learning rate is trained to prove the validity of a transaction and hence, improve fraud detection in electronic banking. The proposed method shows promising results compared with logistic regression and support vector machines.


Articles in Press, Accepted Manuscript
Available Online from 30 July 2021
  • Receive Date: 19 April 2020
  • Revise Date: 01 December 2020
  • Accept Date: 22 July 2021