Detecting manager’s fraud risk using text analysis: evidence from Iran

Abstract

Purpose In accounting and finance, researchers have used many ways to detect manager’s fraud risk. Until now, many researchers have used some data mining methods in these two fields to detect this risk. The purpose of this paper is to compare the precision of two data mining methods in detecting such a risk. Design/methodology/approach For this purpose, this paper analyzed the texts of board’s reports and used two methods including the convex optimization (CVX) method and least absolute shrinkage and selection operator (LASSO) regression method. In this way, the words of these reports, which have the greatest power in explaining the manager’s high fraud risk index, were identified. Using these words, this paper presented a model that could detect manager’s high fraud risk index in companies. Findings The results indicated that both methods can detect the manager’s high fraud risk index with a precision between 82.55 and 91.25 percent. The LASSO method was significantly more precise than the CVX method. Research limitations/implications The lack of access to an official and reliable list of firms suspected to fraud and the lack of access to the Microsoft Word (MS Word) file of board’s reports were two of the most important limitations of this study. Practical implications Regulatory bodies and independent auditors can consider the proposed methods in this study for assessing the fraud risk for a firm or other legal parties. Originality/value This paper avoided using merely financial statements data to detect the manager’s fraud risk index and focused on texts of board’s reports for the detection process. The capabilities of data mining and text mining methods for detecting the manager’s fraud risk index using board’s reports were tested in this paper. By comparing CVX and LASSO results, this paper indicated that methods with a binary-dependent variable have more power and are more precise than methods with continuous-dependent variables for detecting fraud.

Using a Text Mining Approach to Hear Voices of Customers from Social Media toward the Fast-Food Restaurant Industry

“…Previous researchers have used LASSO with many successful applications for feature selection. Rahrovi Dastjerdi, Foroghi, and Kiani [33] employed LASSO detect manager's fraud risk, and they found LASSO was more precise than CVX. Choudhury et al [82] applied LASSO to predict cracks in a beam-like structure.…”

“…Many feature selection methods have been well-developed. Among them, least absolute shrinkage and selection operator (LASSO) has been widely and successfully applied in many domains, such as unrevisit intension factors discovery [32], fraud risk detection [33], bankruptcy prediction [34], power prediction [35], and financial ratio selection [36]. Moreover, LASSO can improve prediction accuracy.…”

Enabling Public Security Text-Based Analytics: A Survey to Outline Research Directions

“…Information extracted from other areas, such as clinical registers, that may be combined with police reports of sexual or domestic violence [42,67,68,87] or from managerial and financial reports to support the detection of fraud [49,[105][106][107] are necessary for the enrichment of the analytical process. Combining clinical and police reports can help health professionals understand the causes of trauma that generate psychological illness in victims, supporting a more adjusted treatment, and the police to understand sexual or domestic violence patterns to prevent the occurrence or recurrence of related situations.…”

Notas explicativas explicam? análise da comunicação do gerenciamento de risco a partir de técnicas de text mining

“…Esta crítica conduz para a segunda direção, foco deste estudo, que conecta o dado em texto contábil com a área de mineração de texto. Ainda que ambas correntes contribuam para maior acessibilidade e compreensão destes textos, nesta direção existe o objetivo de utilizar o texto para realizar análises e inferências atuais, como preço das ações e previsão de performance e fraude (Loughran & McDonald, 2016;Rahrovi et al, 2019). Desta forma, mais do que criar um léxico ou um conjunto linguístico específico da contabilidade, as técnicas de mineração de texto podem auxiliar a tornar este dado mais presente nas avaliações contábeis para gestores e stakeholders.…”