Comparison Analysis of K-Means and DBSCAN Algorithms for Improving Budget Absorption Efficiency in EIS

Abstract

This study aims to analyze the comparison between the K-Means and DBSCAN clustering algorithms in budget absorption within the Executive Information System (EIS). Realized budget achievement data from regional devices serve as the primary dataset for analysis. Before conducting experiments, the data undergo a preprocessing stage to eliminate outliers and apply normalization processes, ensuring the data is ready for further analysis. Subsequently, both algorithms, K-Means and DBSCAN, are applied to the budget achievement data to generate clusters corresponding to their respective characteristics. The research anticipates that the results will unveil significant findings in comparing the performance of K-Means and DBSCAN algorithms in budget absorption within the EIS context, especially within the scope of this study. Therefore, this analysis is expected to provide valuable insights for stakeholders aiming to enhance the efficiency and effectiveness of budget management through the optimal utilization of clustering algorithms within the EIS.