The increasing number of divorce cases at the Simalungun Religious Court has resulted in a growing volume of case data. However, the existing Case Tracking Information System (SIPP) is primarily used for administrative purposes and does not support the analysis of divorce case patterns. This study aims to implement the K-Medoids algorithm in a web-based information system to cluster divorce cases and present the analysis results visually. The study used 2,232 divorce case records from 2024–2025, consisting of case type, age, district, occupation, education, reason for divorce, and case year. The research stages included data preprocessing, Label Encoding, Min-Max Scaling, clustering using the K-Medoids algorithm, and cluster quality evaluation using the Davies-Bouldin Index (DBI). The results showed that the optimal number of clusters was K=3 with the lowest DBI value of 1.249, while the clustering evaluation produced a DBI value of 1.378. The analysis indicated that all clusters were dominated by the factor of continuous disputes and conflicts, with Cluster 2 containing the largest number of cases. The developed system is capable of presenting cluster visualizations, cluster characteristic analyses, and clustering reports to support data-driven decision-making at the Simalungun Religious Court.
Copyrights © 2026