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

Sukanto Bhattacharya, Dongming Xu, Kuldeep Kumar

Research output: Contribution to journalArticleResearchpeer-review

13 Citations (Scopus)

Abstract

While there is a growing professional interest on the application of Benford's law and "digit analysis" in financial fraud detection, there has been relatively little academic research to demonstrate its efficacy as a decision support tool in the context of an analytical review procedure pertaining to a financial audit. We conduct a numerical study using a genetically optimized artificial neural network. Building on an earlier work by others of a similar nature, we assess the benefits of Benford's law as a useful classifier in segregating naturally occurring (i.e. non-concocted) numbers from those that are made up. Alongside the frequency of the first and second significant digits and their mean and standard deviation, a posited set of 'non-digit' input variables categorized as "information theoretic", "distance-based" and "goodness-of-fit" measures, help to minimize the critical classification errors that can lead to an audit failure. We come up with the optimal network structure for every instance corresponding to a 3 × 3 Manipulation-Involvement matrix that is drawn to depict the different combinations of the level of sophistication in data manipulation by the perpetrators of a financial fraud and also the extent of collusive involvement.

Original languageEnglish
Pages (from-to)576-584
Number of pages9
JournalDecision Support Systems
Volume50
Issue number3
Early online date18 Aug 2010
DOIs
Publication statusPublished - Feb 2011

Fingerprint

Fraud
Decision support systems
Financial Audit
Classifiers
Neural networks
Research
Benford's law
Financial fraud
Manipulation
Auditors
Audit
Artificial neural network
Fraud detection
Goodness of fit
Audit failure
Decision support
Efficacy
Sophistication
Network structure
Academic research

Cite this

Bhattacharya, Sukanto ; Xu, Dongming ; Kumar, Kuldeep. / An ANN-based auditor decision support system using Benford's law. In: Decision Support Systems. 2011 ; Vol. 50, No. 3. pp. 576-584.
@article{d30c1982dc3945beadf1360c6555038a,
title = "An ANN-based auditor decision support system using Benford's law",
abstract = "While there is a growing professional interest on the application of Benford's law and {"}digit analysis{"} in financial fraud detection, there has been relatively little academic research to demonstrate its efficacy as a decision support tool in the context of an analytical review procedure pertaining to a financial audit. We conduct a numerical study using a genetically optimized artificial neural network. Building on an earlier work by others of a similar nature, we assess the benefits of Benford's law as a useful classifier in segregating naturally occurring (i.e. non-concocted) numbers from those that are made up. Alongside the frequency of the first and second significant digits and their mean and standard deviation, a posited set of 'non-digit' input variables categorized as {"}information theoretic{"}, {"}distance-based{"} and {"}goodness-of-fit{"} measures, help to minimize the critical classification errors that can lead to an audit failure. We come up with the optimal network structure for every instance corresponding to a 3 × 3 Manipulation-Involvement matrix that is drawn to depict the different combinations of the level of sophistication in data manipulation by the perpetrators of a financial fraud and also the extent of collusive involvement.",
author = "Sukanto Bhattacharya and Dongming Xu and Kuldeep Kumar",
year = "2011",
month = "2",
doi = "10.1016/j.dss.2010.08.011",
language = "English",
volume = "50",
pages = "576--584",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "3",

}

An ANN-based auditor decision support system using Benford's law. / Bhattacharya, Sukanto; Xu, Dongming; Kumar, Kuldeep.

In: Decision Support Systems, Vol. 50, No. 3, 02.2011, p. 576-584.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - An ANN-based auditor decision support system using Benford's law

AU - Bhattacharya, Sukanto

AU - Xu, Dongming

AU - Kumar, Kuldeep

PY - 2011/2

Y1 - 2011/2

N2 - While there is a growing professional interest on the application of Benford's law and "digit analysis" in financial fraud detection, there has been relatively little academic research to demonstrate its efficacy as a decision support tool in the context of an analytical review procedure pertaining to a financial audit. We conduct a numerical study using a genetically optimized artificial neural network. Building on an earlier work by others of a similar nature, we assess the benefits of Benford's law as a useful classifier in segregating naturally occurring (i.e. non-concocted) numbers from those that are made up. Alongside the frequency of the first and second significant digits and their mean and standard deviation, a posited set of 'non-digit' input variables categorized as "information theoretic", "distance-based" and "goodness-of-fit" measures, help to minimize the critical classification errors that can lead to an audit failure. We come up with the optimal network structure for every instance corresponding to a 3 × 3 Manipulation-Involvement matrix that is drawn to depict the different combinations of the level of sophistication in data manipulation by the perpetrators of a financial fraud and also the extent of collusive involvement.

AB - While there is a growing professional interest on the application of Benford's law and "digit analysis" in financial fraud detection, there has been relatively little academic research to demonstrate its efficacy as a decision support tool in the context of an analytical review procedure pertaining to a financial audit. We conduct a numerical study using a genetically optimized artificial neural network. Building on an earlier work by others of a similar nature, we assess the benefits of Benford's law as a useful classifier in segregating naturally occurring (i.e. non-concocted) numbers from those that are made up. Alongside the frequency of the first and second significant digits and their mean and standard deviation, a posited set of 'non-digit' input variables categorized as "information theoretic", "distance-based" and "goodness-of-fit" measures, help to minimize the critical classification errors that can lead to an audit failure. We come up with the optimal network structure for every instance corresponding to a 3 × 3 Manipulation-Involvement matrix that is drawn to depict the different combinations of the level of sophistication in data manipulation by the perpetrators of a financial fraud and also the extent of collusive involvement.

UR - http://www.scopus.com/inward/record.url?scp=78651073836&partnerID=8YFLogxK

U2 - 10.1016/j.dss.2010.08.011

DO - 10.1016/j.dss.2010.08.011

M3 - Article

VL - 50

SP - 576

EP - 584

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 3

ER -