Utilization of Data Mining Techniques to Detect and Predict Accounting Fraud: A Comparison of Neural Networks and Discriminant Analysis

Utilization of Data Mining Techniques to Detect and Predict Accounting Fraud: A Comparison of Neural Networks and Discriminant Analysis

James A. Rodger (Indiana University of Pennsylvania, USA)
DOI: 10.4018/978-1-59140-057-8.ch009
OnDemand PDF Download:
$37.50

Abstract

Accounting information systems enable the process of internal control and external auditing to provide a first-line defense in detecting fraud (Turpen & Messina, 1997). There are few valid indicators at either the individual or the organizational level which are reliable indicators of fraud prevention (Groveman, 1995). Recent studies have shown that it is nearly impossible to predict fraud. In fact, many of the characteristics associated with white-collar criminals are precisely the traits which organizations look for when hiring employees (Lord, 1997). This paper proposes the use of information systems to deal with fraud through proactive information collection, data mining, and decision support activities.

Complete Chapter List

Search this Book:
Reset