FİNANSAL TABLO HİLESİ RİSKİ TAŞIYAN ŞİRKETLERİN VERİ MADENCİLİĞİ İLE BELİRLENMESİ

Öz

Finansal tablo hilesi, şirketlerin finansal tablolarındaki verileri kendi çıkarları doğrultusunda değiştirerek yayınlamalarıdır. Kurumlara, paydaşlara ve ekonomik yapıya ciddi zararlar veren finansal tablo hilelerinin tespit edilmesi önemli bir problemdir. Bunun için çeşitli denetim mekanizmaları bulunmaktadır. Ancak zaman içerisinde geliştirilebilecek hile yöntemlerine karşı yenilikçi denetim yöntemlerine ihtiyaç duyulmaktadır. Veri madenciliği, finansal tablo hilelerinin tespitinde umut vadeden bir alandır. Veri madenciliğinin sınıflandırma analizinde sınıflandırma metotlarıyla mevcut verilerden örüntüler elde edilir ve bunlar görülmemiş birimlerin sınıflandırılmasında kullanılır. Bu çalışmada veri madenciliğinin sınıflandırma metotları ile finansal tablo hilesi riski taşıyan şirketlerin tespiti üzerine bir araştırma yapılmıştır. Veriler Borsa İstanbul’da 2014-2018 arasında işlem gören şirketlerin yayınladıkları finansal tablolardan elde edilmiştir. İlk olarak yedi sınıflandırılma metodu kullanılmış, en başarılı üçü seçilmiştir. Sonraki aşamada başarım değerlerinin geliştirilmesi amacıyla hiper parametre optimizasyonu yapılmıştır. Sınıflandırma metotlarından K-Nearest Neighbor ile yüzde 91,73, Random Forest ile yüzde 90,51 ve XGBoost ile yüzde 90,37 doğruluk oranlarına ulaşılmış, en iyi tahmin oranı K-Nearest Neighbor ile elde edilmiştir. Son kısımda rasgele alt örnekleme yöntemiyle yapılan karşılaştırmalarda da en iyi performans değerleri K-Nearest Neighbor ile elde edilmiştir.

Anahtar Kelimeler

• Finansal Tablo Hilesi
• Veri Madenciliği
• Sınıflandırma

IDENTIFYING THE COMPANIES WITH THE RISK OF FINANCIAL STATEMENT FRAUD BY DATA MINING

Öz

Financial statement fraud is when companies change and publish the data in their financial statements in line with their interests. Detecting financial statement fraud that causes serious damage to organizations, stakeholders, and the economic structure is an important problem. There are various control mechanisms for this. However, there is a need for innovative control methods against new fraud methods that may be developed over time. Data mining is a promising field for detecting financial statement fraud. In the classification analysis of data mining, patterns are obtained from existing data with classification methods, and these are used in the classification of unseen units. In this study, research was carried out on the identification of companies with the risk of financial statement fraud through the classification methods of data mining. The data were obtained from the financial statements published by companies traded in Borsa Istanbul between 2014 and 2018. Seven classification methods were used first, and the three most successful ones were selected. In the next stage, hyperparameter optimization was carried out to improve the performance values. Accuracy rates of 91.73 percent with K-Nearest Neighbor, 90.51 percent with Random Forest, and 90.37 percent with XGBoost were obtained from the classification methods, and the best prediction rate was obtained with K-Nearest Neighbor. In the last part, the best performance values were also obtained with K-Nearest Neighbor in the comparisons using random sub-sampling method.

Anahtar Kelimeler

• Financial Statement Fraud
• Data Mining
• Classification

Kaynakça

1. Abdioğlu, H. (2007). Hilelerin Önlemesi ve Ortaya Çıkarılmasına Yönelik Proaktif Yaklaşımlar. Muhasebe ve Denetime Bakış, 22, 119-137.
2. ACFE. (2020). Report to The Nations: 2020 Global Study on Occupational Fraud and Abuse. Association of Certified Fraud Examiners: https://www.acfe.com/report-to-the-nations/2020/
3. Ata, H. A. & Seyrek, I. H. (2009). The Use of Data Mining Techniques in Detecting Fraudulent Financial Statements: An Application on Manufacturing Firms. Suleyman Demirel University Journal of Faculty of Economics & Administrative Sciences, 14(2), 157-170.
4. Bai, B., Yen, J. & Yang, X. (2008). False Financial Statements: Characteristics of China's Listed Companies and CART Detecting Approach. International Journal of Information Technology & Decision Making, 7(2), 339-359. doi:Doi 10.1142/S0219622008002958
5. Brownlee, J. (2017). What is the Difference Between a Parameter and a Hyperparameter? https://machinelearningmastery.com/difference-between-a-parameter-and-a-hyperparameter/
6. Cotton, D. L. (2002). Fixing CPA Ethics Can Be an Inside Job. https://www.washingtonpost.com/archive/opinions/2002/10/20/fixing-cpa-ethics-can-be-an-inside-job/b7441564-e0a6-431b-9280-8c27c6267ebc/
7. Deng, Q. & Mei, G. (2009). Combining Self-Organizing Map and K-means Clustering for Detecting Fraudulent Financial Statements. 2009 IEEE International Conference on Granular Computing.
8. Erol, M. (2008). İşletmelerde Yaşanan Yolsuzluklara (Hata Ve Hileler) Karşı Denetimden Beklentiler. Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 13(1), 229-237.

9. Ertikin, K. (2017). Hile Denetimi: Kırmızı Bayrakların Tespiti İçin Kullanılan Proaktif Yaklaşımlar. Muhasebe ve Finansman Dergisi, 75, 71-93.
10. Gaganis, C. (2009). Classification Techniques for the Identification of Falsified Financial Statements: A Comparative Analysis. Intelligent Systems in Accounting, Finance & Management: International Journal, 16(3), 207-229.
11. Granitto, P. M., Furlanello, C., Biasioli, F. & Gasperi, F. (2006). Recursive Feature Elimination with Random Forest For ptr-ms Analysis of Agroindustrial Products. Chemometrics and Intelligent Laboratory Systems, 83(2), 83-90. doi:https://doi.org/10.1016/j.chemolab.2006.01.007
12. Hajek, P. & Henriques, R. (2017). Mining Corporate Annual Reports for Intelligent Detection of Financial Statement Fraud – A Comparative Study of Machine Learning Methods. Knowledge-Based Systems, 128, 139-152. doi:10.1016/j.knosys.2017.05.001
13. Han, J., Kamber, M. & Pei, J. (2011). Data Mining: Concepts and Techniques. Elsevier.
14. Hatunoğlu, Z., Koca, N. & Kıllı M. (2012). İç Kontrolün Muhasebe Sistemindeki Hata Ve Hilelerin Önlenmesindeki Rolü Üzerine Bir Alan Çalışması. Mustafa Kemal Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 9(20), 169-189.
15. Hoogs, B., Kiehl, T., Lacomb, C. & Senturk, D. (2007). A Genetic Algorithm Approach to Detecting Temporal Patterns Indicative of Financial Statement Fraud. Intelligent Systems in Accounting, Finance & Management: International Journal, 15(1‐2), 41-56.
16. İşgüden Kılıç, B., Anadolu, Z. (2018). Dijital Çağın Yarattığı Muhasebe Uygulamalarının Muhasebe Hilelerinin Önlenmesine Etkisi. Muhasebe ve Vergi Uygulamaları Dergisi, Özel Sayı: 55-97.
17. Jan, C.-l. (2018). An Effective Financial Statements Fraud Detection Model for the Sustainable Development of Financial Markets: Evidence from Taiwan. Sustainability, 10(2), 513.
18. Kirkos, E., Spathis, C. & Manolopoulos, Y. (2007). Data Mining Techniques for the Detection of Fraudulent Financial Statements. Expert Systems with Applications, 32(4), 995-1003. doi:https://doi.org/10.1016/j.eswa.2006.02.016
19. Kotsiantis, S., Koumanakos, E., Tzelepis, D. & Tampakas, V. (2006). Forecasting Fraudulent Financial Statements using Data Mining. International Journal of Computational Intelligence, 3(2), 104-110.
20. Lenard, M. J., Watkins, A. L. & Alam, P. (2007). Effective Use of Integrated Decision Making: An Advanced Technology Model for Evaluating Fraud in Service-Based Computer and Technology Firms. Journal of Emerging Technologies in Accounting, 4(1), 123-137.
21. 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. Knowledge-Based Systems, 89, 459-470.
22. Lipton, Z. C., Elkan, C. & Naryanaswamy, B. (2014). Optimal Thresholding of Classifiers to Maximize F1 Measure, Berlin, Heidelberg.
23. Perols, J. (2011). Financial Statement Fraud Detection: An Analysis of Statistical and Machine Learning Algorithms. Auditing: A Journal of Practice & Theory, 30(2), 19-50. doi:10.2308/ajpt-50009
24. Ravisankar, P., Ravi, V., Rao, G. R. & Bose, I. (2011). Detection of Financial Statement Fraud and Feature Selection Using Data Mining Techniques. Decision Support Systems, 50(2), 491-500.
25. Rezaee, Z. (2005). Causes, Consequences, and Deterence of Financial Statement Fraud. Critical Perspectives on Accounting, 16(3), 277-298.
26. Rizki, A. A., Surjandari, I. & Wayasti, R. A. (2017). Data Mining Application to Detect Financial Fraud in Indonesia's Public Companies. 3rd International Conference on Science in Information Technology (ICSITech).
27. Scikit-Learn. (2020a). sklearn.feature_selection.RFE. https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html
28. Scikit-Learn. (2020b). sklearn.preprocessing.RobustScaler. https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.RobustScaler.html
29. Sjardin, B., Massaron, L. & Boschetti, A. (2016). Large Scale Machine Learning with Python. Packt Publishing, ISBN: 9781785887215.
30. Terzi, S. & Şen, İ. K. (2012). Finansal Tablo Hilelerinin Veri Madenciliği Yardımıyla Tespit Edilmesi: Üretim Sektöründe Bir Araştırma. Determination of Fraudulent Financial Statements Using Data Mining: A Research in Manufacturing Sector, 5(2), 25-40.
31. Terzi, S. & Şen, İ. K. (2015). Adli Muhasebede Hilelerin Tespitinde Yapay Sinir Ağı Modelinin Kullanımı. International Journal of Economic and Administrative Studies, 7(14), 477-490.
32. Thara, D. K., PremaSudha, B. G. & Xiong, F. (2019). Auto-Detection of Epileptic Seizure Events Using Deep Neural Network with Different Feature Scaling Techniques. Pattern Recognition Letters, 128, 544-550.
33. Uğurlu, M. & Sevim, Ş. (2015).Finansal Tablolardaki Hile Riskinin Tahmin Edilmesinde Karma Modellerin Nispi Başarısı Üzerine Karşılaştırmalı Bir Analiz. Gaziantep University Journal of Social Sciences, 14(1), 65-88.
34. Ulucan Özkul.F & Pertekin, P. (2009). Muhasebe Yolsuzluklarının Tespitinde Adli Muhasebecinin Rolü ve Veri Madenciliği Tekniklerinin Kullanılması. MÖDAV, 2009/4, 57-88.
35. Witten, I. H., Frank, E. & Hall, M. A. (2011). Data Mining: Practical Machine Learning Tools and Techniques (3rd ed.). Burlington, MA: Morgan Kaufmann.
36. Yao, J., Zhang, J. & Wang, L. (2018). A Financial Statement Fraud Detection Model Based on Hybrid Data Mining Methods. International Conference on Artificial Intelligence and Big Data (ICAIBD).