Data Mining Techniques for Auditing Attest Function and Fraud Detection

Data mining technique is a newly developed tool for statisticians, data analysts, and the management information systems community. It involves searching information through databases for correlations and other non-random patterns. In making business decisions, it is important to recognize patterns of data and relationships among variables in order to discover valuable information. The results will best minimize costs, maximize returns, and promote operating efficiency. In the field of accounting and auditing, there is a vast amount of data accumulated in electronic form, and therefore data mining technique is proving to be extremely useful. It allows accountants to analyze the data in many different ways. It can sort through the data, summarize the relationship and reveal the information that the accountants need. This paper explores some applications of data mining techniques as an auditing tool, fraud detection scheme and as an instrument for investigating improper payments. It also compares the general auditing software with the data mining software, for the purpose of showing the superiority of the modern data mining technology. This paper further offers guidance to auditors in the use of data mining software. * The authors are, respectively, Professor of Operations Research and Professor of Accounting at Montclair State University.