An Application Of A Data Mining Technique For Assessing The Risk Of Management Fraud

Abstract

Data mining techniques identify relationships, patterns, trends, and predictive information form large and complex databases. This study demonstrates the use of a data mining technique to assess the risk of management fraud. We use a data mining tool to analyze the management fraud data, presence or absence of red flags in fraud and no fraud cases, collected by a Big Six firm. The ensuing results compare favorably with the statistical and neural network results obtained by the other studies. The study illustrates the ease of using data mining techniques by demonstrating the rapid development of models, querying capabilities, and ease of encoding statistical models in audit decision making.