<![CDATA[Indonesia faces a persistent digital divide between urban and rural areas despite significant progress in internet penetration. This disparity, driven by unequal infrastructure and socio-economic differences, highlights the need for a data-driven approach to infrastructure planning. This study proposes an integrated spatial recommendation model to identify and prioritize regions for internet infrastructure development. The methodology integrates multiple spatial analysis techniques, including optimized K-Means and DBSCAN clustering for socio-economic and infrastructure segmentation, Kernel Density Estimation (KDE) for identifying demand hotspots, and logistic regression and spatial regression as proxies for network accessibility and potential analysis. The analytical results are aggregated into a weighted composite score that classifies each region into high, medium, or low priority categories. The evaluation results show that the optimized K-Means model achieved a Silhouette Score of 0.631, DBSCAN scored 0.766, regression achieved R² = 0.68, and network accessibility classification reached an accuracy of 0.733 with an F1-score of 0.692. The final spatial map successfully identifies high-priority areas across provinces, revealing that infrastructure needs are locally distributed rather than regionally uniform. This framework provides an evidence-based decision-making tool for policymakers and telecommunication providers to guide targeted and equitable investments in digital infrastructure across Indonesia.]]>
