AI-Driven Risk Analytics for Real-Time Fraud Detection in Modernized U.S. Payment Ecosystems
International Journal of Scientific Research in Science, Engineering and Technology · 23.11.2025 · 12 (6) · 586-599
DOI: 10.32628/ijsrset2613147 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Scientific Research in Science, Engineering and Technology, online erschienen am 23.11.2025. Verfasst von Ezekiel Fatomilola. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study investigates the application of artificial intelligence (AI)-driven risk analytics for real-time fraud detection within modernized United States payment ecosystems. As digital payment infrastructures transition toward instant, high-volume transaction processing, traditional rule-based systems exhibit critical limitations, including high false-positive rates and poor adaptability to evolving fraud typologies. This research synthesizes recent advancements in machine learning, deep learning, and hybrid analytical frameworks to evaluate their effectiveness in enhancing detection accuracy, reducing latency, and improving risk decisioning. A systematic review methodology is employed, integrating empirical findings from peer-reviewed literature, fintech implementations, and regulatory analyses. The study identifies that ensemble learning models and deep neural architectures significantly outperform legacy systems, achieving detection accuracy exceeding 95% while maintaining real-time processing capabilities. Additionally, the integration of federated learning and explainable AI (XAI) addresses key challenges related to data privacy, regulatory compliance, and model transparency. The findings demonstrate that a hybrid architecture combining AI-driven predictive analytics with human-in-the-loop governance provides a scalable and resilient fraud mitigation framework. The paper further proposes strategic and operational recommendations for financial institutions, emphasizing interoperability, data quality management, and compliance alignment. This research contributes to the evolving discourse on intelligent financial security by presenting a unified, governance-aware AI risk analytics framework for next-generation payment systems.
Zitieren
APA
Fatomilola, E. (2025). AI-Driven Risk Analytics for Real-Time Fraud Detection in Modernized U.S. Payment Ecosystems. International Journal of Scientific Research in Science, Engineering and Technology, 12(6), 586-599. https://doi.org/10.32628/ijsrset2613147
Vancouver
Fatomilola E. AI-Driven Risk Analytics for Real-Time Fraud Detection in Modernized U.S. Payment Ecosystems. International Journal of Scientific Research in Science, Engineering and Technology. 2025;12(6):586-599. doi:10.32628/ijsrset2613147
BibTeX
@article{Fatomilola2025,
author = {Fatomilola, Ezekiel},
title = {AI-Driven Risk Analytics for Real-Time Fraud Detection in Modernized U.S. Payment Ecosystems},
journal = {International Journal of Scientific Research in Science, Engineering and Technology},
year = {2025},
volume = {12},
number = {6},
pages = {586--599},
doi = {10.32628/ijsrset2613147}
}
EUVIMED