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

The leading digit distribution of the worldwide illicit financial flows

“…However, Benford's law is not definitive [35] in the sense a deviation from it does not prove manipulation just as the conformity to it does not prove the truthfullness and further research is necessary before making any final decision on the quality of the data. Nevertheless Benford's law is useful in analytical procedures for testing the completeness of financial reports [36,37]. Benford's law has been proved to successfully unravel falsification of financial documents, tax evasion by individuals, manipulated trade invoices and tax returns submitted by the companies [15,38], the illicit practices which significantly contribute to the IFFs [32].…”

Detecting Fraudulent Financial Information of a Company Using Hidden Markov Model

“…Notable ones among these works include a study of the relation between the board of director composition and financial information fraud [2], a study of the relation between fraud type and auditor litigation [3] and another that throw light on the link between earnings and operating cash flows and incidence of financial reporting fraud [4]. In the real-life world, the series data of financial ratios often have some nonlinear features, then, ANNs is chosen as our primary tool to perform better in fitting these situations as they are a nonlinear, nonparametric function [5] [6], for examples Green and Choi [7], Lin et al [8], Fanning and Cogger [9] and Feroz et al [10]. Of course, ANNs is not an unmixed blessing methodology.…”

“…A potential drawback of ANNs lies in the use of simulated rather than actual financial data. Unfortunately, it is extremely difficult to obtain enough quantity of real-life data that contains some degree of "contamination" (see [11] [12]). In contrast, HMM is used in a new way to detect the outliers in many areas such as in the works [13] [14], T. Nguyen et al [13] use HMM to discover the key features for the earthquake generation which are not accessible to direct observation, I. Votsi et al [14] introduce an approach to cancer classification through gene expression profiles by designing supervised learning hidden Markov models (HMMs).…”

Text mining to detect indications of fraud in annual reports worldwide

“…In the subsequent years, the financial ratios formed the input for machine learning techniques, such as Decision Trees, Neural Networks and Bayesian Belief Networks, K-nearest neighbors and Support Vector Machines (SVM) (Kotsiantis et al, 2006;Kirkos et al, 2007). In particular, neural networks were a popular machine learning method to detect fraud based on financial information (Green and Choi, 1997;Fanning and Cogger, 1998;Bhattacharya et al, 2011;Huang et al, 2012).…”

“…The newest and popular stateof-the-art machine learning technique is deep learning. Deep learning models use neural networks containing more than one hidden layer, opposed to the neural networks popular for experimentation with financial information to detect fraud (Green and Choi, 1997;Fanning and Cogger, 1998;Bhattacharya et al, 2011;Huang et al, 2012). The deep learning models are successfully used for various types of text analysis research tasks.…”