Network Fraud Detection Using Artificial Neural Networks

¹Ikechukwu F. C. Onah, ²H. C. Inyiama

Department of Computer Engineering, Enugu State University of Science & Technology, Enugu

Department of Electronics and Computer Engineering, Nnamdi Azikiwe University, Awka

E-mail: ikonah@yahoo.co.uk, hcinyiama2002@yahoo.com 

ABSTRACT

The constantly changing nature of network attacks requires a flexible defensive system that is capable of analyzing the enormous amount of network traffic in a manner which is less structured than rule-based systems. In this research paper, the analytical strengths of Artificial Neural Networks have been proposed to identify the typical characteristics of system users and determine statistically significant variations from the user’s established behaviour. The advantages and limitations of neural nets are presented. The paper went on to explain the training process and learning paradigms of Artificial Neural Networks. An Artificial Neural Network agent can be deployed in a multi-agent architecture for the purpose of observing, gathering and recording data that can be used in detecting frauds within a network.

Keywords: Fraud detection, neural networks, Intrusion detection, Fraud classifiers.