This document offers an innovative perspective on optimizing the detection of irregularities in public resource management by integrating Bayesian models, fuzzy logic, and large-scale data analysis. Through an empirical study focused on the case of Veracruz (2011–2016), it demonstrates the value of combining probabilistic inference techniques with the flexibility of fuzzy logic to classify the severity of accounting and financial findings. In addition, the inclusion of machine learning and Big Data algorithms provides enhanced capabilities for analyzing large volumes of information, identifying anomalous patterns, and generating early alerts.
Its multidisciplinary approach reveals how traditional audits can be strengthened through mathematical and computational tools, offering greater accuracy and speed in governmental oversight. Readers will find a practical proposal for prioritizing risks and irregularities, enabling more effective allocation of oversight efforts. With these contributions, the work stands out as an essential resource for auditors, researchers, and public officials seeking to innovate in transparency and accountability, ensuring the correct and efficient use of public funds.
- Quote paper
- Carlos Medel-Ramírez (Author), 2025, Aplicación de Modelos Bayesianos y Lógica Difusa para la Detección Automatizada de Irregularidades en la Fiscalización Gubernamental, Munich, GRIN Verlag, https://www.grin.com/document/1572553
