<![CDATA[The integration of the K-Means Clustering algorithm and the Random Forest Regressor aims to map divorce patterns in Indonesia for the 2018–2024 period. This study introduces a paradigm for categorizing variables into axis factors—such as economic issues, disputes, and abandonment—and triggering factors, which can be referred to as behavioral factors, such as apostasy, gambling, domestic violence, and so on. The clustering results show a Silhouette Score of 0.5912, with Java Island identified as the region with the highest risk and a well-defined risk zoning. The Random Forest model achieved an R-squared accuracy of 0.9563, revealing that trigger factors—specifically gambling, incarceration, apostasy, and domestic violence—possess a higher deterministic weight in precipitating divorce compared to other factors. This key finding indicates that the impact/triggering factors do not always correlate directly with the number of cases. Therefore, it recommends policy interventions focused on breaking the chain of behavioral triggers to strengthen family resilience nationwide.]]>
Copyrights © 2026
