Detection of Financial Statement Fraud Using Deep Learning for Sustainable Development of Capital Markets under Information Asymmetry

Jan, Chyan-Long

doi:10.3390/su13179879

Cited by 29 publications

(13 citation statements)

References 38 publications

(56 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normally, words in English can be easily recognized since the space token is a good approximation of a word divider. Different from English (or more broadly, languages that use some form of the Latin alphabet), there are no interval marks between words in Chinese (or other languages that do not have obvious word delimiters such as Korean and Japanese) [38]. Therefore, it is difficult for word segmentation to identify ambiguous words in Chinese document preprocessing.…”

Section: A Chinese Text Embeddingmentioning

confidence: 99%

An Analysis on Financial Statement Fraud Detection for Chinese Listed Companies Using Deep Learning

2022

IEEE Access

View full text Add to dashboard Cite

Financial fraud has extremely damaged the sustainable growth of financial markets as a serious problem worldwide. Nevertheless, it is fairly challenging to identify frauds with highly imbalanced dataset because ratio of non-fraud companies is very high compared to fraudulent ones. Intelligent financial statement fraud detection systems have therefore been developed to support decision-making for the stakeholders. However, most of current approaches only considered the quantitative part of the financial statement ratios while there has been less usage of the textual information for classifying, especially those related comments in Chinese. As such, this paper aims to develop an enhanced system for detecting financial fraud using a state-of-the-art deep learning models based on combination of numerical features that derived from financial statement and textual data in managerial comments of 5130 Chinese listed companies' annual reports. First, we construct financial index system including both financial and non-financial indices that previous researches usually excluded. Then the textual features in MD&A section of Chinese listed company's annual report are extracted using word vector. After that, powerful deep learning models are employed and their performances are compared with numeric data, textual data and combination of them, respectively. The empirical results show great performance improvement of the proposed deep learning methods against traditional machine learning methods, and LSTM, GRU approaches work with testing samples in a correct classification rate of 94.98% and 94.62%, indicating that the extracted textual features of MD&A section exhibit promising classification results and substantially reinforce financial fraud detection.

show abstract

Section: A Chinese Text Embeddingmentioning

confidence: 99%

An Analysis on Financial Statement Fraud Detection for Chinese Listed Companies Using Deep Learning

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Various deep-learning algorithms have also been employed in financial fraud identification research. Examples include convolutional neural networks (CNNs) [ 27 ], long short-term memory (LSTM) [ 28 ], hierarchical self-attention (HSA) [ 29 ], and self-organizing maps (SOMs) [ 30 ]. However, single-classifier models are limited by the models themselves, and their performance improvement has become a bottleneck; hence, many scholars have turned their attention to ensemble-learning algorithms.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Many scholars have always selected financial ratio indicators [ 41 , 42 ] as fraud risk identification indicators when studying financial fraud identification, and some scholars have included nonfinancial indicators such as ownership structure [ 28 ] in the model indicator input. Few scholars have considered introducing textual information in annual reports into the indicators, and only a few scholars have attempted to do so.…”

Section: Literature Reviewmentioning

confidence: 99%

Financial Fraud Identification Based on Stacking Ensemble Learning Algorithm: Introducing MD&A Text Information

Zhang

Yong

Yong-jun

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In recent years, there have been frequent incidents of financial fraud committed through various means. How to more efficiently identify financial fraud and maintain capital market order is a problem that scholars from all walks of life are discussing and urgently seeking to resolve. In this study, a financial fraud identification model is constructed based on the stacking ensemble learning algorithm, and the text of the management discussion and analysis (MD&A) chapter in annual reports is introduced based on financial and nonfinancial variables, using sentiment polarity, emotional tone, and text readability as text variables. The results show that when considering financial and nonfinancial variables and introducing text variables, the recognition effect of the stacking ensemble learning model constructed in this study is significantly better than the classification results of each single classifier model. In addition, the model recognition effect is better after adding text variables. Therefore, the model is expected to provide a new and more effective method of identifying financial fraud.

show abstract

“…In addition to great efforts to prevent their occurrence, fraud in financial statements occurs from time to time (Jan, 2021), with the risk of their occurrence increased by globalization of business, market growth and rapid technological development (Jan, 2018). Fraud is a global problem, i.e., it is present in all countries of the world (Luković & Stojković, 2020).…”

Section: Introductionmentioning

confidence: 99%

“…The emergence of the COVID-19 pandemic, in addition to a large number of infected and deceased persons, also led to rising unemployment, large financial losses and endangered functioning of the capital market (Jan, 2021). The issue of ensuring business stability, sustainability and long-term growth, which has been called into question by the emergence of the COVID-19 pandemic (International Federation of Accountants, 2020), may encourage individuals to resort to fraudulent financial reporting (Deloitte, 2020).…”

Section: Introductionmentioning

confidence: 99%

Covid-19 Pandemic and Fraudulent Actions in the Financial Statements: The Case of Hotel Companies in the Republic of Serbia

Milašinović

Jovković

Dragičević³

2022

View full text Add to dashboard Cite

The quality of information presented in the financial statements of companies may be impaired due to fraudulent activities. One of the frequently used instruments that indicates that fraud has been committed in the financial statements is the Beneish M-score model. The purpose of the research is to examine the exposure of hotel companies in the Republic of Serbia to the risk of fraud in financial statements, as well as to determine whether there has been a change in that risk due to the COVID-19 pandemic. The research was conducted on a sample of 100 randomly selected hotel companies, and their financial reports for 2019 and 2020 were observed. There was a high risk of fraud based on the recognition of costs of sales, management and administration, and borrowing in both observed years, as well as a high risk of fraud based on the recognition of income and gross margin index in 2020. Further, it was determined that in the year of the COVID-19 pandemic, there was an increase in the risk of fraud in the financial statements based on the gross margin index and recognition of selling, management and administration costs. Microenterprises are exposed to a higher risk of fraud based on the recognition of costs of sales, management and administration compared to medium-sized ones. Limited liability companies are more exposed to risk based on the recognition of sales revenue, accrual item and total risk (measured by the 8-variable model) compared to joint-stock companies. On the other hand, joint-stock companies are more exposed to the risk of fraud based on gross margin and general risk of fraud (measured by the 5-variable model) compared to limited liability companies.

show abstract

Detection of Financial Statement Fraud Using Deep Learning for Sustainable Development of Capital Markets under Information Asymmetry

Cited by 29 publications

References 38 publications

An Analysis on Financial Statement Fraud Detection for Chinese Listed Companies Using Deep Learning

An Analysis on Financial Statement Fraud Detection for Chinese Listed Companies Using Deep Learning

Financial Fraud Identification Based on Stacking Ensemble Learning Algorithm: Introducing MD&A Text Information

Covid-19 Pandemic and Fraudulent Actions in the Financial Statements: The Case of Hotel Companies in the Republic of Serbia

Contact Info

Product

Resources

About