Data-Driven Transparency: Machine Learning and Social Network Analysis for Corruption Detection in Public Procurement

Corruption in public procurement is a global challenge that can cause inefficiencies, financial losses, and a decline in public trust. Fraudulent behaviors, such as supplier collusion or conflicts of interest, undermine the integrity, transparency, and integrity of procurement processes. Traditional monitoring mechanisms struggle to detect these situations due to the complexity and volume of procurement processes. This paper proposes a methodology to detect corrupt behavior in public procurement using Machine Learning and Social Network Analysis. Our model is trained to identify suspicious patterns in supplier relationships, improving due diligence and risk assessment. As a case study, we apply our methodology to Chilean public procurement data, demonstrating its potential to improve transparency and mitigate corruption risks.