Detecting Fraudulent Financial Statements Using Deep Learning Algorithms

Subject Areas : Ethics and accounting

Javad Sharafkhani ¹ , Ali EsamailZadeh Mogry ² , Mohammad Ali Bidari ³

1 - PhD., Student, Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran
2 - Professor, Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran
3 - Assistant Professor, Department of Accounting, Central Tehran Branch, Islamic Azad University, Tehran, Iran.

Received: 2025-01-11 Accepted : 2025-02-14 Published : 2025-04-22

Keywords: Fraudulent Financial Statements, Fraud Detection, Deep Learning, Recurrent Neural Network Algorithms,

Abstract :

Objective: The purpose of this study is to detect fraudulent financial statements in companies listed on the Tehran Stock Exchange using deep learning algorithms.

Method: The study analyzed 1,800 company-years (150 companies over 12 years) of annual financial reports from companies listed on the Tehran Stock Exchange, covering the period from 2012 to 2023. Three deep learning algorithms were employed in this research: support vector machine (SVM), convolutional neural network (CNN), and recurrent neural network (RNN). Additionally, the "two-sample mean comparison test" method was employed to select the final variables for the study and to develop the model.

Findings: The findings indicate that the overall accuracy of the deep learning algorithms is as follows: Support Vector Machine (SVM) at 88.4%, Convolutional Neural Network (CNN) at 81.3%, and Recurrent Neural Network (RNN) at 94.4%. This suggests that the RNN algorithm demonstrates the best performance, while the CNN algorithm exhibits the lowest performance in detecting companies with fraudulent financial statements. In other words, the results indicate
that the recurrent neural network (RNN) algorithm is more efficient than other deep learning algorithms. Therefore, among the three algorithms—Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN)—used in companies listed on the Tehran Stock Exchange, the RNN algorithm offers the most effective model for detecting fraudulent financial statements.

Conclusion: The findings of this study offer valuable insights for shareholders, creditors, analysts, and other participants in the capital markets. These insights can enhance the prediction of fraudulent financial statements, minimize errors in decision-making based on financial information, improve the evaluation of company performance, facilitate the adoption of optimal trading strategies, and help identify suitable opportunities for buying and selling stocks based on financial statement data.

References:

احمدی، سید جلال؛ فغانی ماکرانی، خسرو؛ فاضلی، نقی (1399). ارائه مدلی برای پیش‌بینی صورت‎های ‌مالی متقلبانه و مقایسه صورت‌ها و نسبت‌های مالی با قانون بنفورد. دانش حسابداری و حسابرسی مدیریت، 9(35)، ص221-237.
احمدی، سید جلال؛ فغانی ماکرانی، خسرو؛ فاضلی، نقی (1403). تکنیک‌های داده‌کاوی و پیش‌بینی تقلب صورت‌های مالی. دانش حسابداری و حسابرسی مدیریت، 13(52)، ص15-28.
اعتمادی، حسین؛ زلقی، حسن (1392). کاربرد رگرسیون لجستیک در شناسایی گزارشگری مالی متقلبانه. دانش حسابرسی، 13(51)، ص26-1.
امیرمعزی، حسین؛ پورآقاجان، عباسعلی؛ جعفری، علی (1403). کشف تقلب صورت‌های مالی: قیاس توانایی مدل‌های مبتنی بر متغیرهای حسابداری. دانش حسابداری و حسابرسی مدیریت، 13(52)، ص173-188.
بهشتی مسئله‌گو، سیده مژگان؛ افشار کاظمی، محمدعلی؛ حقیقت منفرد، جلال؛ رضاییان، علی (1401). یادگیری عمیق برای پیش‌بینی بازار سهام با استفاده از اطلاعات عددی و متنی (رویکرد الگوریتم حافظه کوتاه مدت ماندگار. مهندسی مالی و مدیریت اوراق بهادار (پذیرفته شده برای انتشار).
خواجوی، شکرالله؛ ابراهیمی، مهرداد (1396). ارائۀ یک رویکرد محاسباتی نوین برای پیش‌بینی تقلب در صورت‌های مالی، با استفاده از شیوه‌های خوشه‌بندی و طبقه‌بندی (شواهدی از شرکت‌های پذیرفته‌شده در بورس اوراق بهادار تهران). پیشرفت‌های حسابداری، 9(2)، ص1-34.
رضائی، مهدی؛ ناظمی اردکانی، مهدی؛ ناصر صدرآبادی، علیرضا (1400). پیش‌بینی تقلب صورت‌های مالی با استفاده از رویکرد کریسپ (CRISP). دانش حسابداری و حسابرسی مدیریت، 10(40)، ص150-135.
رضائی، مهدی؛ ناظمی اردکانی، مهدی؛ ناصر صدرآبادی، علیرضا (1400). پیش‌بینی تقلب صورت‌های مالی با استفاده از رویکرد کریسپ(CRISP). دانش حسابداری و حسابرسی مدیریت، 10(40)، ص135-150.
زارعی، علی؛ رهنمای رودپشتی، فریدون؛ خان‌محمدی، محمدحامد؛ کردلویی، حمیدرضا (1403). ارائه الگوی پیش‌بینی تقلب مبتنی بر هوش مصنوعی (بیز ساده). مطالعات اخلاق و رفتار در حسابداری و حسابرسی، 4(4)، ص7-26.
سیاهکارزاده، علی‌محمد (1400). پیش‌بینی بازارهای مالی با استفاده از یادگیری عمیق. پایان‌نامه کارشناسی ارشد، مهندسی کامپیوتر. دانشگاه صنعتی شاهرود.
صفرزاده، محمدحسین (1389). توانایی نسبت‌های مالی در کشف تقلب در گزارشگری مالی: تحلیل لاجیت. دانش حسابداری، 1(1)،ص 137-163.
کاظمی، توحید؛ پیری، پرویز (1401). پیش‌بینی طرح تقلب در گزارشگری مالی با استفاده از رویکرد یادگیری ماشین در فضای چند کلاسه. پژوهش‌های تجربی حسابداری، 12(46)، ص276-255.
ملکی کاکلر، حسن؛ بحری ثالث، جمال؛ جبارزاده کنگرلویی، سعید؛ آشتاب، علی (1400). کارایی مدل‌های آماری و الگوهای یادگیری ماشین در پیش‌بینی گزارشگری مالی متقلبانه. اقتصاد مالی، 15(54)، ص267-292.
Albashrawi, M. (2016). Detecting financial fraud using data mining techniques: J. Data Sci. no.14,
p. 553–569.
Alfaiz, N.S. & Fati, S.M. (2022). Enhanced Credit Card Fraud Detection Model Using Machine Learning. Electronics, no.11, p. 662.
Alghofaili, Y., Albattah, A., Rassam, M. & Rassam, M. (2020). A Financial Fraud Detection Model Based on LSTM Deep Learning Technique. Journal of Applied Security Research, 15(4),
p. 498–516. https://doi.org/10.1080/19361610.2020.1815491.
Ali, A.A., Khedr, A.M., El-Bannany, M. & Kanakkayil, S. (2023). A Powerful Predicting Model for Financial Statement Fraud Based on Optimized XGBoost Ensemble Learning Technique. Appl. Sci.no. 13, p. 2272. https://doi.org/10.3390/ app13042272
Alsinglawi, M.M.A.S.M.A.O. & Almari, M.O.S. (2021). Predicting Fraudulent Financial Statements Using Fraud Detection Models. Acad. Strateg. Manag, no. 20, p.1–17.
Amar, I.A.A.B. & Jarboui, A. (2013). Detection of Fraud in Financial Statements: French Companies as a Case Study. Int. J. Acad. Res. Bus. Soc. Sci. no. 3, p. 456–472.
Andrew, C. & Robin, C (2022). Detecting Fraudulent of Financial Statements Using Fraud S.C.O.R.E Model and Financial Distress. Int. J. Econ. Bus. Account. Res. no. 6, p. 211–222.
Asimit, A.V., Kyriakou, I., Santoni, S., Scognamiglio, S. & Zhu, R. (2022). Robust Classification via Support Vector Machines. Risks, no.10, p.154.
Cecchini, M., Aytug, H., Koehler, G.J. & Pathak, P. (2010). Detecting management fraud in public companies. Manag. Sci. no. 56, p.1146–1160.
Chen, S. (2016). Detection of fraudulent financial statements using the hybrid data mining approach. SpringerPlus, no. 5, p. 1–16.
Craja, P., Kim, A. & Lessmann, S. (2020). Deep learning for detecting financial statement fraud. Decis. Support Syst, no.139, p.113421.
Dechow, P., Ge, W., Larson, C. & Sloan, R. (2011). Predicting Material Misstatements. Contemporary Accounting Research, 28(1), p. 17-82.
Deebak, B., Memon, F.H., Dev, K., Khowaja, S.A., Wang, W. & Qureshi, N.M.F. (2022).TAB-SAPP: A trust-aware blockchain-based seamless authentication for massive IoT-enabled industrial applications. IEEE Trans. Ind. Inform. No. 19, p. 243250.
El-Bannany, M., Sreedharan, M. & Khedr, A.M. (2020). A robust deep learning model for financial distress prediction. Int. J. Adv. Comput. Sci. Appl. No. 11, p.170–175.
Gupta, R. & Gill, N.S. (2012). Prevention and detection of financial statement fraud–An implementation of data mining framework. Editor. Pref. no 3, p.150–160
Gupta, S. & Mehta, S.K. (2021). Data mining-based financial statement fraud detection: Systematic literature review and meta-analysis to estimate data sample mapping of fraudulent companies against non-fraudulent companies. Glob. Bus. Rev., 25(2), p. 1–26.
Hamal, S. & Senvar, Ö. (2021). Comparing performances and effectiveness of machine learning classifiers in detecting financial accounting fraud for Turkish SMEs. Int. J. Comput. Intell. Syst. No. 14, p. 769–782.
Han, D. (2017). Researches of Detection of Fraudulent Financial Statements Based on Data Mining.
J. Comput. Theor. Nanosci., no. 14, p. 32–36.
Humpherys, S.L., Moffitt, K.C., Burns, M.B., Burgoon, J.K. & Felix, W.F. (2011). Identification of fraudulent financial statements using linguistic credibility analysis. Decis. Support Syst. No. 50,
p. 585–594.
Jan, Ch. (2021). Detection of Financial Statement Fraud Using Deep Learning for Sustainable Development of Capital Markets under Information Asymmetry. Sustainability 13(17), p. 9879.
Katsis, D.Ch. & et al. (2012). Using Ants to Detect Fraudulent Financial Statements. Journal of Applied Finance & Banking, 2(6), p. 73-81.
Kulikova, L. & Satdarova, D. (2016). Internal control and compliance-control as effective methods of management, detection and prevention of financial statement fraud. Acad. Strateg. Manag. J. no. 15, p. 92.
Kumar, R. & Tripathi, R. (2020). Secure healthcare framework using blockchain and public key cryptography. In: Blockchain Cybersecurity, Trust and Privacy; Springer: Cham, Switzerland,
p. 185–202.
Kumar, S., Ahmed, R., Bharany, S., Shuaib, M., Ahmad, T., Tag Eldin, E., Rehman, A.U. & Shafiq, M. (2022). Exploitation of Machine Learning Algorithms for Detecting Financial Crimes Based on Customers’ Behavior. Sustainability, no.14, p. 13875.
Lin, C.C., Chiu, A.A., Huang, S.Y. & Yen, D.C. (2015). Detecting the financial statement fraud: The analysis of the differences between data mining techniques and experts’ judgments. Knowl-Based Syst, no. 89, p. 459–470.
Moepya, S.O., Akhoury, S.S. & Nelwamondo, F.V. (2014). Cost-sensitive classification for financial fraud detection under high classimbalance. In: Proceedings of the 2014 IEEE international conference on data mining workshop, Shenzhen, China, 14–17 December 2014; IEEE: New York, NY, USA, p. 183–192.
Mohammadi, M., Yazdani, S., Khanmohammadi, M.H. & Maham, K. (2020). Financial reporting fraud detection: An analysis of data mining algorithms. Int. J. Financ. Manag. Account, no. 4,
p. 1–12.
Murorunkwere, B.F., Tuyishimire, O., Haughton, D. & Nzabanita, J. (2022). Fraud Detection Using Neural Networks: A Case Study of Income Tax. Future Internet, no.14, p. 168.
Pai, P.F., Hsu, M.F. & Wang, M.C. (2011). A support vector machine-based model for detecting top management fraud. Knowl-Based Syst, no. 24, p. 314–321.
Paulo Sérgio Gomes Macedo, H.C.I. & Vieira, E.S. (2022). A model to detect financial statement fraud in Portuguese companies by the auditor. Contaduría Adm, no. 67, p. 185–209.
Pérez López, C., Delgado Rodríguez, M. & de Lucas Santos, S. Tax Fraud Detection through Neural Networks: An Application Using a Sample of Personal Income Taxpayers. Future Internet,
no. 11, p. 86.
Perols, J. (2011). Financial statement fraud detection: An analysis of statistical and machine learning algorithms. Audit. J. Pract. Theory, no. 30, p. 19–50.
Perols, J. (2011). Financial Statement Fraud Detection: An Analysis of Statistical and Machine Learning Algorithms. A Journal of Practice & Theory, 30(2), p. 19-50.
Ravisankar, P., Ravi, V., Rao, G.R. & Bose, I. (2011). Detection of financial statement fraud and feature selection using data mining techniques. Decis. Support Syst. No. 50, p. 491–500.
Rizki, A.A., Surjandari, I. & Wayasti, R.A. (2017). Data mining application to detect financial fraud in Indonesia’s public companies. In: Proceedings of the 2017 3rd International Conference on Science in Information Technology (ICSITech), Bandung, Indonesia, 25–26 October 2017; IEEE: New York, NY, USA, p. 206–211.
Schreiber-Gregory, D. & Bader, K. (2018). Logistic and Linear Regression Assumptions: Violation Recognition and Control. In: Proceedings of the SESUG Conference, St. Pete Beach, FL, USA,
p. 1–6.
Spathis, C., Doumpos, M. & Zopounidis, C. (2002). Detecting falsified financial statements: A comparative study using multicriteria analysis and multivariate statistical techniques. European Accounting Review, 11(3), p. 509-535.
Sreedharan, M., Khedr, A.M. & El Bannany, M. (2020). A Multi-Layer Perceptron Approach to Financial Distress Prediction with Genetic Algorithm. Autom. Control. Comput. Sci. no. 54,
p. 475–482.
Zaheer, S., Anjum, N., Hussain, S., Algarni, A.D., Iqbal, J., Bourouis, S. & Ullah, S.S. (2023). A Multi Parameter Forecasting for Stock Time Series Data Using LSTM and Deep Learning Model. Mathematics, no.11, p. 590. https://doi.org/10.3390/ math11030590