<![CDATA[The increasing complexity of corporate crime has heightened the need for consistent prosecutorial practices to ensure legal certainty, fairness, and accountability within the criminal justice system. However, empirical studies examining the consistency of prosecutorial demands in corporate criminal cases remain limited, particularly those employing data-driven analytical approaches. This study aims to analyze the consistency patterns of prosecutors' demands in corporate criminal cases using court decision data and a Machine Learning approach. The research employed a quantitative design based on data-driven legal analytics, utilizing corporate criminal court decisions as the unit of analysis. A Random Forest algorithm was developed to identify demand patterns based on case characteristics, including the type of offense, financial loss, corporate benefit, management involvement, recovery efforts, and aggravating or mitigating circumstances. The findings indicate that the proposed model achieved high predictive performance, demonstrating that prosecutorial demands generally follow consistent patterns influenced primarily by the magnitude of financial loss, the type of offense, and the economic benefits obtained by the corporation. These findings highlight the potential of Machine Learning to provide an objective framework for evaluating prosecutorial consistency and supporting evidence-based prosecutorial policies. This study contributes to the advancement of computational legal studies by introducing a data-driven framework for assessing prosecutorial discretion in corporate criminal cases and offers practical implications for improving transparency, accountability, and consistency in prosecution practices.]]>
Copyrights © 2026
