2020· International Journal of Commerce, Finance and Digital Economy· 0 citations
TL;DR
It is concluded that AI is a critical component of modern financial security infrastructure and will play an increasingly important role in combating financial fraud.
Abstract
The rapid growth of digital banking, e-commerce, electronic payments, and financial technology has increased both the volume and complexity of financial transactions, leading to greater fraud risks. Traditional rule-based fraud detection systems struggle to identify evolving and sophisticated fraud patterns. Artificial Intelligence (AI) offers an effective solution through machine learning, pattern recognition, and predictive analytics. This study explores AI-based fraud prevention systems in financial services, focusing on data acquisition, preprocessing, feature engineering, classification, anomaly detection, and real-time monitoring. Various machine learning models, including Random Forest, Support Vector Machines, Artificial Neural Networks, and Deep Learning, are evaluated for fraud detection. The findings indicate that AI-driven systems significantly improve fraud detection accuracy, reduce false positives, minimize operational losses, and enhance customer trust. The study concludes that AI is a critical component of modern financial security infrastructure and will play an increasingly important role in combating financial fraud.
The findings demonstrate that AI-driven fraud monitoring represents a fundamental component of modern financial security infrastructure and will continue to shape the future of fraud prevention in increasingly digital financial environments.
G. Onyarin· International Journal For Mu...· 0 citations
The review highlights the transformative potential of AI while emphasizing the need for ethical, transparent, and secure implementation strategies to maximize effectiveness in combating increasingly sophisticated financial fraud schemes.
G. Onyarin· International journal of res...· 0 citations
The rapid digitalisation of financial services has created enormous opportunities for banks, financial institutions and businesses, but it has also increased their exposure to fraud and financial risks. Conventional fraud detection systems, which mainly depend on predefined rules and manual investigation, are increasingly challenged by sophisticated and rapidly changing fraudulent activities. Artificial Intelligence (AI), particularly Machine Learning (ML), Deep Learning, Natural Language Processing and anomaly detection, provides financial institutions with new ways to identify suspicious patterns, assess risks and respond to potential threats in real time. This paper examines the role of AI in fraud detection and financial risk management, with particular attention to its applications in banking and digital financial services. The study is conceptual and descriptive in nature and is based on primary data and secondary information obtained from academic literature and reports of financial and international institutions. The paper discusses how AI can improve transaction monitoring, credit risk assessment, anti-money-laundering activities, cybersecurity and predictive risk management. The study argues that AI should not be viewed as a complete replacement for human judgement. Rather, the most effective approach is likely to combine AI-driven analysis with human expertise, appropriate governance and continuous monitoring. The paper concludes that responsible and explainable AI can significantly strengthen the ability of financial institutions to prevent fraud and manage financial risks while maintaining customer trust and financial stability.
V. Soniya· International journal of res...· 0 citations
Digital payment systems have become the backbone of global commerce, but their rapid expansion has been paralleled by a sharp rise in payment fraud, identity theft, and cyber-enabled financial crime. This paper examines the role of Artificial Intelligence (AI) in enhancing the security and fraud-detection capability of digital online payment systems, drawing on recent industry reports, regulatory data, and empirical machine learning studies. The study adopts a descriptive-analytical approach, synthesising secondary data from central bank publications, market-research reports, and peer-reviewed comparative studies of algorithms such as Random Forest, Artificial Neural Networks, Support Vector Machines, and Gradient Boosting Results show that AI-powered fraud detection tools, such as the ensemble method Random Forest, consistently outperform their human counterparts, with accuracy between 92 and 100 percent in experimental and production environments; real-time behavioural analytics, biometric authentication and natural language processing also take fraud protection beyond transaction-level screening to include phishing, social engineering and mule-account detection. The scale of the challenge, as well as the regulatory response to the menace of digital payment fraud, is evident from the Indian Unified Payments Interface (UPI) ecosystem, where the value of digital payment fraud fluctuated despite an over 40 per cent increase in the number of transactions year-on-year, and the Reserve Bank of India's (RBI) MuleHunter.AI initiative. However, the paper still points to certain issues that have not been overcome, such as class imbalance, manipulation by a growing swarm of AI-savvy fraudsters, a lack of explanation, data-privacy restrictions, and disparity in adoption by institutions of varying sizes. The paper concludes that AI plays an essential role in the current payment-security architecture, but it must be complemented with a multi-layered approach that includes technological solutions, regulations, and consumer-awareness initiatives to effectively withstand the ever-changing threat landscape.
Ch. Keerthi, B. Nandini· Advanced International Journ...· 0 citations
Data quality, class imbalance, privacy preservation, model interpretability, scalability and regulatory compliance are among the key challenges that are critically analyzed.
Amit Jain· International Journal of Adv...· 0 citations
Experimental findings show that hybrid models are much more effective than standalone classifiers with respect to precision, recall, F 1 -score and area under the ROC curve (AUC) particularly in detecting rare and unseen cases of frauds.
M. Mohammed· International Journal of App...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
