EVOLUSI DETEKSI KECURANGAN: TINJAUAN LITERATUR SISTEMATIS PADA DOMAIN KEUANGAN, PERPAJAKAN, DAN BLOCKCHAIN

Abstract

This study aims to examine the evolution of fraud detection methods across three strategic domains: finance, taxation, and blockchain. This topic is significant because fraud not only causes financial losses but also undermines public trust and threatens the stability of economic systems. The hypothesis of this study suggests that technology-based approaches, such as machine learning and data analytics, improve fraud detection effectiveness compared to conventional methods. A Systematic Literature Review (SLR) method was employed, analyzing 6.294 primary scientific articles published between 2019 and 2025. The analysis was conducted thematically and bibliometrically to identify key trends and research gaps. The findings reveal a paradigm shift in fraud detection, from behavior-based theoretical models to algorithm-driven analytical systems. However, the integration of technological approaches with behavioral and organizational factors remains limited. This study provides a comprehensive overview of current fraud detection developments and highlights the need for a multidisciplinary approach that combines technical, behavioral, and institutional dimensions. These findings are essential as a foundation for strengthening fraud surveillance and governance systems in the digital era.