An ESG-Compliant Framework for Fraud Detection in Online Payments Using Data Privacy and Machine Learning

Purpose: The purpose of this paper is to analyze the role of machine learning (ML) in fraudulent online payment transaction detection across industries, with a specific focus on integrating Environmental, Social, and Governance (ESG) principles to bring sustainability and ethics in fraud detection. Methodology: This study reviews the effectiveness of machine learning (ML) techniques in real-time fraud detection, presenting the efficiency of different ML techniques across industries and how those techniques can be modified to apply the principles of ESG. It checks on ethical practices concerning data, compliance with data privacy, and sustainable management, further investigating how effectively ML-driven frameworks can support fraud detection without compromising the standards of ESG. Findings: The study finds that integrating ESG principles within machine learning frameworks for fraud detection enhances both the effectiveness and ethical alignment of these systems. ML models not only support real-time fraud detection but also reinforce sustainable data management and governance practices, providing businesses with an advanced approach to mitigating cyber risks while upholding ESG commitments. Unique Contribution to Theory, Practice, and Policy: This paper contributes by advancing ESG-integrated frameworks for fraud detection, offering a sustainable and ethical model for using ML in financial systems. In practice, it provides actionable insights for industries seeking to align fraud prevention with ESG goals. The paper also suggests policy considerations, advocating for stronger ethical standards and data governance in ML applications for fraud detection.