Document Type : Research Article

Authors

• Bagher Shamloo ¹
• somaye parhizkary ²

¹ Professor, Department of Criminal Law and Criminology, Faculty of Law, Shahid Beheshti University, Tehran, Iran

² PHD student of criminal law and criminology, Faculty of Law, Shahid Beheshti University, Tehran, Iran.

Abstract

The data-based approach has been introduced in recent years as a new paradigm in the prevention of economic crimes. The transition from traditional rule-based and risk-based systems to systems based on big data, alternative data, and machine learning algorithms has made it possible to carryout more efficiently the processes of customer due diligence and the detection of suspicious transactions—two fundamental pillars emphasized by international instruments to combat economic crimes. The present article, employing a descriptive-analytical method and aims to identify the capacities and challenges of applying this approach. The research findings indicate that machine learning algorithms, in both supervised and unsupervised branches, demonstrate considerable capability in uncovering hidden patterns and anomalies in financial transactions, and reduce the rates of false positives and false negatives. Conversely, the application of this approach faces challenges such as algorithmic opacity and its conflict with the right of defense, as well as the vulnerability of algorithmic evidence to adversarial attacks. It is concluded that although the data-based approach has opened a clear pathway for the prevention of economic crimes, particularly money laundering, the full realization of these capacities will not be possible without due consideration given to the design of legal frameworks and intelligent regulation

Keywords

• data-based approach
• prevention
• customer due diligence
• suspicious transactions
• machine learning
• algorithmic transparency

Main Subjects

• Criminal Law