Mining corporate annual reports for intelligent detection of financial statement fraud - A comparative study of machine learning methods

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216275%3A25410%2F17%3A39910599" target="_blank" >RIV/00216275:25410/17:39910599 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.sciencedirect.com/science/article/pii/S0950705117302022" target="_blank" >http://www.sciencedirect.com/science/article/pii/S0950705117302022</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.knosys.2017.05.001" target="_blank" >10.1016/j.knosys.2017.05.001</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Mining corporate annual reports for intelligent detection of financial statement fraud - A comparative study of machine learning methods

Popis výsledku v původním jazyce

Financial statement fraud has been serious concern for investors, audit firms, government regulators, and other capital market stakeholders. Intelligent financial statement fraud detection systems have therefore been developed to support decision-making of the stakeholders. Fraudulent misrepresentation of financial statements in managerial comments has been noticed in recent studies. As such, the purpose of this study was to examine whether an improved financial fraud detection system could be developed by combining specific features derived from financial information and managerial comments in corporate annual reports. To develop this system, we employed both intelligent feature selection and classification using a wide range of machine learning methods. We found that ensemble methods outperformed the remaining methods in terms of true positive rate (fraudulent firms correctly classified as fraudulent). In contrast, Bayesian belief networks (BBN) performed best on non-fraudulent firms (true negative rate). This finding is important because interpretable &quot;green flag&quot; values (for which fraud is likely absent) could be derived, providing potential decision support to auditors during client selection or audit planning. We also observe that both financial statements and text in annual reports can be utilised to detect non-fraudulent firms. However, non-annual report data (analysts&apos; forecasts of revenues and earnings) are necessary to detect fraudulent firms. This finding has important implications for selecting variables when developing early warning systems of financial statement fraud.

Název v anglickém jazyce

Mining corporate annual reports for intelligent detection of financial statement fraud - A comparative study of machine learning methods

Popis výsledku anglicky

Financial statement fraud has been serious concern for investors, audit firms, government regulators, and other capital market stakeholders. Intelligent financial statement fraud detection systems have therefore been developed to support decision-making of the stakeholders. Fraudulent misrepresentation of financial statements in managerial comments has been noticed in recent studies. As such, the purpose of this study was to examine whether an improved financial fraud detection system could be developed by combining specific features derived from financial information and managerial comments in corporate annual reports. To develop this system, we employed both intelligent feature selection and classification using a wide range of machine learning methods. We found that ensemble methods outperformed the remaining methods in terms of true positive rate (fraudulent firms correctly classified as fraudulent). In contrast, Bayesian belief networks (BBN) performed best on non-fraudulent firms (true negative rate). This finding is important because interpretable &quot;green flag&quot; values (for which fraud is likely absent) could be derived, providing potential decision support to auditors during client selection or audit planning. We also observe that both financial statements and text in annual reports can be utilised to detect non-fraudulent firms. However, non-annual report data (analysts&apos; forecasts of revenues and earnings) are necessary to detect fraudulent firms. This finding has important implications for selecting variables when developing early warning systems of financial statement fraud.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA16-19590S" target="_blank" >GA16-19590S: Analýza témat a sentimentu vícenásobných textových zdrojů pro finanční rozhodování podniků</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Knowledge-Based Systems

ISSN

0950-7051

e-ISSN

—

Svazek periodika

128

Číslo periodika v rámci svazku

July

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

14

Strana od-do

139-152

Kód UT WoS článku

000403633500012

EID výsledku v databázi Scopus

—