<![CDATA[The growing implementation of governance models in higher education institutions has raised new challenges in accountability and performance. Although risk management and analysis in governance has received some attention in the literature, there is a lack of studies focusing on the use data-driven clustering algorithms. The current research focuses on the development and assessment of a web-based risk management information system that incorporates clustering principles to identify and map non-academic governance risks. The study followed the CRISP-DM framework and analyzed 678 risks from the Internal Audit Unit (SPI) of Politeknik Negeri Malang. The system employs K-Means clustering analysis to classify the risks into tiers based on performance indicators, budget, and risk severity. The system is equipped with data upload, preprocessing, logging, and cluster visualization modules. A comparative analysis of K-Means, DBSCAN, and Hierarchical Clustering showed that K-Means yields the best cluster quality with a Silhouette Score of 0.48, in comparison to DBSCAN (-0.705) and Hierarchical clustering (-0.395) . The developed system generated five distinct clusters corresponding to risks of varying settlement priority viz. very high, high, medium, low, and very low. The functional and usability assessments of the system confirmed that it provides automated and actionable insights on risks in a user-centric manner. The study has demonstrated the clustering of governance risks in higher education using K-Means is feasible. The incorporation of predictive analytics and real-time data would best support the research in active risk avoidance.]]>
