A New Prototype for Intelligent Visual Fraud Detection in Agent-Based Auditing Framework

Abstract

While US. Sarbanes Oxley act has been viewed by most as an onerous and expensive requirement; it is having a positive impact on driving appropriate levels of investment in IT security, controls, and transactional systems. This paper introduces a new secure solution for auditing and accounting based on artificial intelligence technology. These days, security is a big issue among regulatory firms. Big companies are concerned about their data to be disseminated to their competitors; this high risk prevents them to provide full information to the regulatory firms. This solution not only significantly reduces the risk of unauthorized access to the company’s information but also facilitate a framework for controlling the flow of disseminating information in a risk free method. Managing security is performed by a network of mobile agents in a pyramid structure among regulatory organization like securities and exchanges commissions, stock exchanges in top of this pyramid to the companies in the button. Because of security considerations, our strategy is to delegate all fraud detection algorithms to Intelligent Mobile Auditing Agent and web service undertake all inter communicational activity. Web services can follow auditing actives in predefined framework and they can act based on permitted security allowance to auditors. The current solution is designed based on Java-based mobile agents. Such design reaps strong mobility and security benefits. This new prototyped solution could be a framework for strengthening security for future development in this area. An insider trading case study is used to demonstrate and evaluate the approach.