<![CDATA[In facing the challenges of limited resources and business complexity, the Investment and One-Stop Integrated Services Office (DPMPTSP) of Langkat Regency requires a data-driven approach to determine priorities for business supervision and enforcement. This study applies the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm to cluster business entities based on three main parameters: risk level, business scale, and licensing status. Secondary data from 3,748 companies were collected, processed through label encoding and normalization, and analyzed in a three-dimensional space (X1_Risk, X2_Scale, X3_License). The clustering results revealed the formation of clusters and a Silhouette Score value, indicating optimal cluster structure and separation between groups. Each cluster was interpreted as a representation of recommendation categories such as Routine Monitoring and Evaluation, Intensive Monitoring and Evaluation, Administrative Warning, Temporary Operational Suspension, and Permanent Operational Termination. The resulting visualizations enhanced the understanding of spatial mapping and clustering patterns comprehensively. This demonstrates that DBSCAN is effective as a decision-support tool for automated and objective priority mapping in business supervision, and capable of detecting business entities that deviate from general norms (outliers). This approach significantly contributes to improving the efficiency and accuracy of decision-making in business license supervision and enforcement at the regional level.]]>
Copyrights © 2025
