<![CDATA[This study evaluates the impact of implementing Machine Learning architectures on strategic decision-making within financial institutions in Metropolitan Lima. Using an experimental approach, models such as XGBoost were developed, achieving an AUC-ROC of 0.92 and an F1-Score of 0.89. The results demonstrate a significant improvement in operational indicators between 2024 (pre-intervention) and 2025 (post-intervention). Specifically, credit delinquency decreased from 4.8% to 4.2%, while fraud losses dropped by 25% (from 0.8% to 0.6%). Additionally, the credit approval time was reduced from 15 to 5 days, contributing to an 18.2% increase in return on assets (ROA). The findings confirmed that transitioning from reactive to proactive predictive models optimizes operational efficiency and financial resilience. This research provides a technical and strategic roadmap for the digital transformation of the banking sector through advanced analytics and a data-driven culture.]]>
Copyrights © 2026
