Abstract

Public procurement remains one of the most vulnerable areas to corruption globally, costing governments billions in lost revenue and eroding public trust. Traditional anti-corruption measures, often reactive and fragmented, fail to address the systemic complexities and dynamic nature of procurement fraud. This study proposes a novel predictive framework grounded in algorithmic integrity to proactively identify, mitigate, and prevent corruption in public procurement using artificial intelligence (AI) and data analytics. By integrating machine learning models with procurement datasets such as bidder profiles, contract values, timelines, and performance metrics the framework can detect anomalous patterns, bid-rigging schemes, and conflicts of interest in real time. Drawing on supervised and unsupervised learning techniques, including decision trees, clustering algorithms, and neural networks, the proposed model delivers a scalable and adaptable mechanism for high-risk contract flagging and fraud prediction. The study also embeds ethical AI principles to ensure fairness, transparency, and accountability in algorithmic decision-making, thereby reinforcing public sector integrity. The framework is validated through a case study involving a national procurement database, revealing significant improvement in early detection of corruption indicators compared to conventional audit-based methods. Results demonstrate over 87% accuracy in identifying suspicious transactions, with false positives minimized through iterative model training and feedback loops. Furthermore, the integration of explainable AI tools enhances trust and interpretability for procurement officers and anti-corruption agencies. This research contributes to the growing discourse on digital governance by offering an evidence-based, technology-driven solution to systemic corruption in public procurement. The findings underscore the importance of data transparency, institutional cooperation, and policy alignment in operationalizing algorithmic tools for public good. Ultimately, the study calls for the adoption of algorithmic integrity frameworks as a standard in digital anti-corruption infrastructure and public financial management reform.