Machine learning – based financial fraud detection: Evidence from healthcare and related firms in Türkiye

Öz

Financial fraud and accounting irregularities distort financial transparency, weaken corporate governance, and reduce the efficiency of resource allocation. Machine learning techniques have emerged as tools for detecting complex fraud-related patterns that traditional approaches may fail to capture. This study examines the effectiveness of machine learning in identifying financial fraud risk among healthcare and healthcare-related firms in Türkiye. Using firm-level data covering 2021–2025 (279 firm-quarter observations, 15 firms), six algorithms — Logistic Regression, Naive Bayes, k-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest — are evaluated through firm-level five-fold cross-validation with hyperparameter tuning and explicit class-imbalance handling. Fraud-risk indicators are derived from the Beneish framework, while predictor variables are kept independent from the fraud-risk label to avoid target leakage. The findings reveal a clear trade-off: Random Forest achieves the highest accuracy and specificity (72.4% and 86.7%) but the lowest sensitivity (21.3%), whereas Decision Tree achieves the highest sensitivity (60.7%) and F-score (0.354) despite lower accuracy (51.6%). No single algorithm dominates across all criteria. The pruned decision tree indicates that changes in gross profit margin and quarterly net profit margin are the signals most consistently associated with elevated fraud risk. The results highlight both the potential and limitations of data-driven approaches for financial monitoring, showing that model selection should depend on the relative cost of false negatives versus false positives. Rather than replacing auditor judgment, the evaluated models are best understood as decision-support tools that help prioritize firms for closer review within a transparent, reproducible framework.

Anahtar Kelimeler

• Machine Learning
• Financial Fraud Detection
• Financial Statement Analysis
• Healthcare Sector
• Beneish M-Score.

Makine öğrenmesi temelli finansal hile tespiti: Türkiye’de sağlık ve ilişkili firmalar üzerine bir uygulama

Öz

Finansal hileler ve muhasebe düzensizlikleri, finansal şeffaflığı zayıflatarak kurumsal yönetişimi olumsuz etkilemekte ve ekonomik kaynakların etkin dağılımını bozabilmektedir. Son yıllarda firma düzeyinde finansal verilerin artan erişilebilirliği, karmaşık hile kalıplarının tespitinde makine öğrenmesi tekniklerinin kullanımını mümkün kılmıştır. Bu çalışma, Türkiye’de faaliyet gösteren sağlık ve sağlıkla ilişkili sektörlerdeki firmalarda finansal hile riskinin belirlenmesinde makine öğrenmesi yöntemlerinin etkinliğini incelemeyi amaçlamaktadır. Analizde 2021–2025 dönemine ait firma düzeyinde finansal veriler (15 firmadan 279 firma-çeyrek gözlem) kullanılmış; Lojistik Regresyon, Naive Bayes, k-En Yakın Komşu, Destek Vektör Makineleri, Karar Ağacı ve Random Forest olmak üzere altı gözetimli makine öğrenmesi algoritması, firma düzeyinde beş katlı çapraz doğrulama, hiperparametre optimizasyonu ve sınıf dengesizliğinin açıkça yönetildiği bir prosedürle değerlendirilmiştir. Hile riski göstergeleri, Beneish modeli temelinde hesaplanan finansal oranlar ve hedef sızıntısını önlemek amacıyla hile riski etiketinden bağımsız tutulan ek finansal değişkenler kullanılarak oluşturulmuştur. Bulgular, modeller arasında açık bir ödünleşim olduğunu göstermektedir: Random Forest en yüksek doğruluk ve özgüllük oranına (sırasıyla %72,4 ve %86,7) ulaşırken en düşük duyarlılığa (%21,3) sahiptir; Karar Ağacı ise daha düşük genel doğruluk oranına (%51,6) rağmen en yüksek duyarlılığa (%60,7) ve F-skoruna (0,354) ulaşmaktadır. Hiçbir algoritma tüm değerlendirme kriterlerinde üstünlük sağlamamaktadır. Budanmış karar ağacı bulguları, brüt kâr marjı değişimi ve çeyreklik net kâr marjı değişiminin bu örneklemde yüksek hile riskiyle en tutarlı şekilde ilişkili finansal sinyaller olduğunu göstermektedir. Sonuçlar, veri odaklı analitik yaklaşımların finansal izleme açısından hem potansiyelini hem de sınırlılıklarını ortaya koymakta; model seçiminin, hedeflenen uygulamada yanlış negatif ile yanlış pozitif hataların göreli maliyetine bağlı olması gerektiğini vurgulamaktadır.

Anahtar Kelimeler

• Makine Öğrenmesi
• Finansal Hile Tespiti
• Finansal Tablo Analizi
• Sağlık Sektörü
• Beneish M-Skoru.

Kaynakça

1. Abbas, M., Almulla, D., Alqashra, A. Y. and Al-Shammari, M. (2024, December). Applying machine learning to detect fraud of financial statements: A systematic literature review. Conference presentation. 2024 International Conference on Decision Aid Sciences and Applications (DASA), Manama, Bahrain. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10836535&isnumber=10836159
2. Agrawal, N. and Panigrahi, S. (2023, May). A comparative analysis of fraud detection in healthcare using data balancing and machine learning techniques. Conference presentation. 2023 International Conference on Communication, Circuits and Systems (IC3S), Bhubaneswar, India. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10169634&isnumber=10169121
3. Ahmed, J. and Ahmed, M. (2022). Machine learning and artificial intelligence in healthcare. In R. Nedunchezhian, M. Rajalakshmi, D. Goyal, S. Balamurugan, A. Elngar and B. Keswani (Eds.), Empowering artificial intelligence through machine learning: New advances and applications (43–56). Apple Academic Press. Access: https://www.appleacademicpress.com/empowering-artificial-intelligence-through-machine-learning-new-advances-and-applications/9781771889308
4. Aksoy, B. (2021). Finansal tablo hilelerinin makine öğrenmesi yöntemleri ve lojistik regresyon kullanılarak tahmin edilmesi: Borsa İstanbul örneği. Maliye ve Finans Yazıları, (115), 29–60. Doi: https://doi.org/10.33203/mfy.733855
5. Alanazi, A. (2022). Using machine learning for healthcare challenges and opportunities. Informatics in Medicine Unlocked, 30(100924), 1–5. Doi: https://doi.org/10.1016/j.imu.2022.100924
6. Amponsah, A. A., Adekoya, A. F. and Weyori, B. A. (2022). A novel detection and prevention method for healthcare claim processing using machine learning and blockchain technology. Decision Analytics Journal, 4(100122), 1–12. Doi: https://doi.org/10.1016/j.dajour.2022.100122
7. Arslankaya, S. and Toprak, Ş. (2021). Makine öğrenmesi ve derin öğrenme algoritmalarını kullanarak hisse senedi fiyat tahmini. Uluslararası Mühendislik Araştırma ve Geliştirme Dergisi, 13(1), 178–192.
8. Bauder, R. A. and Khoshgoftaar, T. M. (2017, December). Medicare fraud detection using machine learning methods. Conference presentation. 16th IEEE International Conference on Machine Learning and Applications (ICMLA), Cancun, Mexico. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8260744&isnumber=8260597

9. Beneish, M. D. (1999). The detection of earnings manipulation. Financial Analysts Journal, 55(5), 24–36. Doi: https://doi.org/10.2469/faj.v55.n5.2296
10. Bolton, R. J. and Hand, D. J. (2002). Statistical fraud detection: A review. Statistical Science, 17(3), 235–255. Doi: https://doi.org/10.1214/ss/1042727940
11. Bounab, R., Guelib, B., Benzerogue, S. and Zarour, K. (2024, November). Optimizing machine learning for healthcare fraud detection: A framework using hybrid feature selection and hyperparameter tuning. Conference presentation. 2024 International Conference on Advanced Aspects of Software Engineering (ICAASE). Constantine, Algeria. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10851054&isnumber=10850921
12. Chen, H., Chiang, R. H. L. and Storey, V. C. (2012). Business intelligence and analytics: From big data to big impact. MIS Quarterly, 36(4), 1165–1188. Doi: https://doi.org/10.2307/41703503
13. Du Preez, A., Bhattacharya, S., Beling, P. and Bowen, E. (2025). Fraud detection in healthcare claims using machine learning: A systematic review. Artificial Intelligence in Medicine, 160(103061), 1–17. Doi: https://doi.org/10.1016/j.artmed.2024.103061
14. Duman, E. A. and Sağıroğlu, Ş. (2017, October). Healthcare fraud detection methods and new approaches. Conference presentation. 2017 International Conference on Computer Science and Engineering (UBMK). Antalya, Turkey. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8093544&isnumber=8093363
15. Fındık, H. and Öztürk, E. (2016). Finansal bilgi manipülasyonunun Beneish modeli yardımıyla ölçülmesi: BIST imalat sanayi üzerine bir araştırma. İşletme Araştırmaları Dergisi, 8(1), 483–499. Doi: https://doi.org/10.20491/isarder.2016.156
16. Kandemir, T. and Kardeş, Z. (2025). Hileli finansal tabloların tespitinde veri madenciliği uygulamaları: Mevcut araştırma eğilimlerinin incelenmesi (2006–2024). Denetişim, (33), 333–355. Doi: https://doi.org/10.58348/denetisim.1674341
17. Kirkos, E., Spathis, C. and 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
18. Köse, İ., Göktürk, M. and Kılıç, K. (2015). An interactive machine-learning-based electronic fraud and abuse detection system in healthcare insurance. Applied Soft Computing, 36, 283–299. Doi: https://doi.org/10.1016/j.asoc.2015.07.018
19. Lavanya, S., Kumar, S. M. and Kumar, P. M. (2021). Machine learning based approaches for healthcare fraud detection: A comparative analysis. Annals of the Romanian Society for Cell Biology, 25(3), 8644–8654. Access: http://annalsofrscb.ro/index.php/journal/article/view/2409
20. Lekkala, L. R. (2023). Importance of machine learning models in healthcare fraud detection. Voice of the Publisher, 9, 207–215. Doi: https://doi.org/10.4236/vp.2023.94017
21. Malhotra, R. and Mishra, V. R. (2024, January). Utilizing machine learning for identification of financial fraud in the healthcare sector. Conference presentation. International Conference on Optimization Computing and Wireless Communication (ICOCWC), Debre Tabor, Ethiopia. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10470779&isnumber=10470444
22. Mehbodniya, A., Alam, I. and Pande, S. (2021). Financial fraud detection in healthcare using machine learning and deep learning techniques. Security and Communication Networks, 2021(9293877), 1–8. Doi: https://doi.org/10.1155/2021/9293877
23. Ngai, E. W. T., Hu, Y., Wong, Y. H., Chen, Y. and Sun, X. (2011). The application of data mining techniques in financial fraud detection: A classification framework and an academic review of literature. Decision Support Systems, 50(3), 559–569. Doi: https://doi.org/10.1016/j.dss.2010.08.006
24. Peker, A. A. (2023). Muhasebe manipülasyonlarının tespit edilmesinde Beneish TR modeli: Borsa İstanbul üzerine sektörel bazda bir uygulama. Muhasebe ve Denetime Bakış Dergisi, (69), 321–346. Doi: https://doi.org/10.55322/mdbakis.1220209
25. 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: https://doi.org/10.2308/ajpt-50009
26. Phua, C., Lee, V., Smith, K. and Gayler, R. (2010). A comprehensive survey of data mining-based fraud detection research. Computer Science, Doi: https://doi.org/10.48550/arXiv.1009.6119
27. Prova, N. N. I. (2024, August). Healthcare fraud detection using machine learning. Conference presentation. 2024 Second International Conference on Intelligent Cyber Physical Systems and Internet of Things (ICoICI 2024), Coimbatore, India. Access: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10696476&isnumber=10695948
28. Ravisankar, P., Ravi, V., Rao, G. R. and Bose, I. (2011). Detection of financial statement fraud and feature selection using data mining techniques. Decision Support Systems, 50(2), 491–500. Doi: https://doi.org/10.1016/j.dss.2010.11.006
29. Shekhar, S., Leder-Luis, J. and Akoglu, L. (2023). Unsupervised machine learning for explainable healthcare fraud detection. National Bureau of Economic Research, w30946, 1–46. Access: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4356230#
30. Waghade, S. S. and Karandikar, A. M. (2018). A comprehensive study of healthcare fraud detection based on machine learning. International Journal of Applied Engineering Research, 13(6), 4175–4178.
31. West, J. and Bhattacharya, M. (2016). Intelligent financial fraud detection: A comprehensive review. Computers & Security, 57, 47–66. Doi: https://doi.org/10.1016/j.cose.2015.09.005
32. Yılmaz, H. (2025). Finansal bilgi manipülasyonunun Beneish TR modeli kullanılarak tahmin edilmesi: BİST imalat sanayi üzerine bir uygulama. Journal of Research in Economics, Politics & Finance, 10(1), 359–388. Doi: https://doi.org/10.30784/epfad.1602273