<![CDATA[Several institutions still face challenges in managing incoming letters due to the absence of a structured classification system. The follow-up process is often carried out based on the order of arrival without considering the urgency or importance of the content, leading to document accumulation and difficulties in determining priorities. To address these issues, this study aims to develop a Decision Support System (DSS) that assists in classifying letters and provides strategic recommendations for follow-up prioritization. The K-Means method is used to cluster letter data based on attribute similarities, supported by Principal Component Analysis (PCA) for dimensionality reduction. Furthermore, the Analytical Hierarchy Process (AHP) is applied to generate strategic recommendations for each cluster of letters. The research data were obtained from the institution’s budget management letter registry, which includes attributes such as institution name, department, description, program, activity, sub-activity, and sub-detail. The results indicate that the K-Means method is less optimal for clustering complex letter data, with a silhouette score of 0.208 and a Davies–Bouldin Index (DBI) of 1.793. However, the AHP method achieved a consistency ratio (CR) below 0.1, demonstrating the reliability of the generated recommendations. Overall, the developed system effectively provides accurate and efficient recommendations for letter follow-up prioritization, thereby improving decision-making processes within the institution.]]>
Copyrights © 2025
