A fuzzy neural network for assessing the risk of fraudulent financial reporting

Abstract

While financial reporting fraud has become more prevalent and costly in recent years, fraud detection has been badly lagging. Several recent studies have examined the feasibility of various computer techniques in business and industrial applications. The purpose of this study is to evaluate the utility of an integrated fuzzy neural network (FNN) for fraud detection. The FNN developed in this research outperformed most statistical models and artificial neural networks (ANN) reported in prior studies. Its performance also compared favorably with a baseline Logit model, especially in the prediction of fraud cases.

An ANN-based auditor decision support system using Benford's law

“…Coakley and Brown [17] addressed some of the deeper modeling issues with ANN applications in accounting and finance. Lin, Hwang and Becker [19] further extended the ANN as a methodological tool in ARP by applying a fuzzy neural network model in the assessment of risk pertaining to fraudulent financial reporting. Kirkos, Spathis and Manolopoulos [12] tested the usefulness of decision trees, neural networks and Bayesian belief networks in identifying fraudulent financial reporting using ratios derived from financial statements to construct the input vectors.…”

The back side of banking in Russia: forecasting bank failures with negative capital

“…They also probably reject bribery as a payment 'to get things done'. Lin and Becker (2003) compare a fuzzy neural network and a logit model for the detecting of fraudulent financial statements issued by US publicly traded companies in 1980-1995. Both models exhibit good forecasting power for non-fraudulent cases, while the fuzzy neural network dominates in classifying frauds.…”

A Framework for Internal Fraud Risk Reduction at IT Integrating Business Processes: The IFR² Framework

“…The studies of Davey et al (1996) and Hilas and Mastorocostas (2008) (telecommunications fraud), Dorronsoro et al (1997) (credit card fraud), and Fanning and Cogger (1998), Green and Choi (1997) and Kirkos et al (2007) (financial statement fraud) all use neural network technology for detecting fraud in different contexts. Lin et al (2003) apply a fuzzy neural net, also in the domain of fraudulent financial reporting. Both Brause et al (1999) and Estévez et al (2006) use a combination of neural nets and rules.…”