AI-Driven Approaches for Real-Time Fraud Detection in US Financial Transactions: Challenges and Opportunities

Bello, Oluwabusayo Adijat and Ogundipe, Abidemi and Mohammed, Damilola and Adebola, Folorunso, and Alonge, Olalekan Ayodeji (2023) AI-Driven Approaches for Real-Time Fraud Detection in US Financial Transactions: Challenges and Opportunities. European Journal of Computer Science and Information Technology, 11 (6). pp. 84-102. ISSN 2054-0957 (Print), 2054-0965 (Online)

This is the latest version of this item.

Text
AI-Driven Approaches.pdf - Published Version
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (526kB) | Request a copy

Official URL: https://eajournals.org/ejcsit/

Abstract

Fraud in financial transactions remains a significant challenge for the US financial sector, necessitating the development of advanced detection mechanisms. Traditional methods, often limited by their reactive nature and inability to handle large volumes of data in real-time, are increasingly being supplemented and replaced by AI-driven approaches. This paper explores the application of artificial intelligence for real-time fraud detection, highlighting the potential benefits, challenges, and future directions of these technologies. AI-driven techniques, such as machine learning algorithms, deep learning models, and natural language processing, offer robust solutions for identifying and mitigating fraudulent activities. Supervised and unsupervised learning methods, alongside anomaly detection techniques, provide the ability to detect unusual patterns and behaviors that may indicate fraud. The integration of hybrid models enhances the accuracy and reliability of these systems. Implementing AI-driven fraud detection systems involves challenges such as ensuring data quality, addressing privacy concerns, and achieving scalability for real-time processing. Additionally, balancing model performance with regulatory compliance and ethical considerations remains a critical concern. Despite these challenges, the advancements in AI technologies present significant opportunities. Enhanced data analytics, collaborative efforts between financial institutions and AI firms, and regulatory support can drive innovation and improve fraud detection capabilities. Case studies from leading financial institutions demonstrate the effectiveness of AI-driven approaches in reducing fraud rates and improving operational efficiency. As AI technology continues to evolve, its application in fraud detection promises a more secure financial environment. This paper provides a comprehensive overview of the current state, challenges, and future potential of AI-driven real-time fraud detection in US financial transactions, aiming to inform and guide stakeholders in the financial sector.

Item Type:	Article
Subjects:	T Technology > T Technology (General)
Depositing User:	Professor Mark T. Owen
Date Deposited:	20 Jun 2024 15:32
Last Modified:	20 Jun 2024 15:32
URI:	https://tudr.org/id/eprint/3127

Available Versions of this Item

AI-Driven Approaches for Real-Time Fraud Detection in US Financial Transactions: Challenges and Opportunities. (deposited 11 Jun 2024 04:26)
- AI-Driven Approaches for Real-Time Fraud Detection in US Financial Transactions: Challenges and Opportunities. (deposited 20 Jun 2024 15:32) [Currently Displayed]

Actions (login required)

: View Item

UNSPECIFIED UNSPECIFIED

CORE (COnnecting REpositories)