European Journal of Economics, Finance and Administrative Sciences
ISSN 1450-2275 Issue 47 (2012)
© EuroJournals, Inc. 2012
http://www.eurojournals.com/EJEFAS.htm
The Possibility of using Artificial Neural Networks in
Auditing -Theoretical Analytical Paper
Reem Okab Hussein Taha
Al, Balqa Applied University, Jordan, Al-Huson College
E-mail: Medoheart2000@yahoo.com
Abstract
This paper is intended to study and evaluate the feasibility of using artificial neural
networks in auditing as research examines the concept of artificial neural networks, the
most important types and the advantages and disadvantages. It shows the most important
differences between expert systems and artificial neural networks. It also examines reasons
why auditing requires to use artificial neural networks and the most important current
models of artificial neural networks applications in auditing.
Most importantly, this paper has found that Fuzzy Neural Network model is better
than conventional statistical methods in planning of auditing process and accomplishment
of auditing task. The neural networks can guide the external auditor to the financial
statements that contains potential substantial errors, in which, therefore, additional auditing
tests should be applied. Fuzzy Neural Network model also assists the external auditor in
predicting the companies whose statements free from manipulation, this helps the external
auditor to give correct opinion on financial statements.
Keywords: Audit. Artificial Neural Networks In Audit
1. Introduction
The spread of using computers in many business units in economic activity has significantly developed
in recent years. Such a development is due to the recent changes occurred in business units , industrial
firms, information technology, computer technology, and the changes in vocational, governmental, and
economic institutions surrounding the firms (Mubarak, Salah al-Din Rifai, Lotfi, 1996, p 3) the most
significant developments in recent years so-called artificial intelligence and its applications in
computers and auditing in particular.
Artificial intelligence is defined as a use of computer in the simulation of human intelligence
via human information (experience)-based computer programs through which a decision can be
reached similar to human decision. However, Decision-Making cannot be based on information only.
Artificial Neural Network is a artificial intelligence application that have begun an interesting topic
since fifties as scientists have tried to make programs that stimulate neural networks function in the
human brain and linked them together in order to make a certain learning process, but they were unable
to do so because the hardware at that time was not capable to process enough quantity of neural
networks to reach something stimulates human intelligence and the brain function was unknown at that
time (. Zain Abdul-Hadi, 2000, p 21),
The interest in artificial neural networks has renewed in late eighties due to the need to process
data in a manner that stimulates human mind, the progress in computer technology and neurology that
resulted in greater understanding of human brain structure (Turban et al., 2001, p. 607). Recently, the
accounting research has tended to use artificial neural networks in dealing with various accounting