A Suite of Interpretable Hybrid Systems with a Voting Mechanism for Detecting Financial Fraud in Digital Forensics
Journal of Al-Farabi for Engineering Sciences · 13.04.2026 · 5 (1) · 42-50
DOI: 10.59746/qr58vk09 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Al-Farabi for Engineering Sciences, online erschienen am 13.04.2026. Verfasst von Alaa Kadum Abed Ali und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Financial fraud has become a critical challenge in modern digital payment systems as financial transactions increasingly rely on online platforms. Despite advancements in payment technologies, detecting fraudulent activities remains difficult due to evolving attack strategies and system vulnerabilities. Traditional fraud detection methods suffer from high false positive rates, inability to adapt to new fraud patterns, and limited interpretability, especially when using complex models that function as black boxes. This study proposes a multilayered fraud detection framework that integrates Decision Tree (DT) and Fuzzy Inference System (FIS) within a unified architecture. Initially, the Synthetic Minority Oversampling Technique (SMOTE) is applied to address class imbalance in the dataset, followed by constructing a Decision Tree model to extract clear and interpretable decision rules. These rules are then transformed into fuzzy logic rules within the Fuzzy Inference System to handle uncertainty and borderline cases more effectively. Finally, a hard voting mechanism is employed to combine the outputs of both models and produce a robust final decision. The proposed model was evaluated using real-world financial transaction data. Experimental results demonstrated an accuracy of 96.75% ± 0.12, outperforming several baseline models such as Random Forest, CNN-SVM, and rule-based systems. The findings indicate that the proposed hybrid framework enhances detection accuracy, reduces false alarms, and improves interpretability, making it a reliable solution for real-time financial fraud detection.
Zitieren
APA
Ali, A. K. A., Majeed, H. L., Gromov, & Y.Y (2026). A Suite of Interpretable Hybrid Systems with a Voting Mechanism for Detecting Financial Fraud in Digital Forensics. Journal of Al-Farabi for Engineering Sciences, 5(1), 42-50. https://doi.org/10.59746/qr58vk09
Vancouver
Ali AKA, Majeed HL, Gromov, Y.Y. A Suite of Interpretable Hybrid Systems with a Voting Mechanism for Detecting Financial Fraud in Digital Forensics. Journal of Al-Farabi for Engineering Sciences. 2026;5(1):42-50. doi:10.59746/qr58vk09
BibTeX
@article{Ali2026,
author = {Ali, Alaa Kadum Abed and Majeed, Huda Lafta and Gromov and Y.Y},
title = {A Suite of Interpretable Hybrid Systems with a Voting Mechanism for Detecting Financial Fraud in Digital Forensics},
journal = {Journal of Al-Farabi for Engineering Sciences},
year = {2026},
volume = {5},
number = {1},
pages = {42--50},
doi = {10.59746/qr58vk09}
}
EUVIMED