| dc.description.abstract | Clustering analysis aims to group areas in Labuhanbatu Regency based on oil palm production using the K-Medoids algorithm, oil palm is one of the important commodities in the national plantation sector that plays a major role in generating foreign exchange and creating jobs. With production results varying between regions, this clustering is expected to help optimize government programs in managing and increasing oil palm production. Cluster analysis using the K-Medoids method was chosen because it is able to group regions based on similar production characteristics. This method uses the Euclidean distance measure to measure the closeness between data, so that regions with similar production results will be grouped in the same cluster. The results show that there are three main clusters based on palm oil production levels. Cluster 1, which represents the low production category, consists of the sub-districts of Bilah Barat, Bilah Hilir, and Pangkatan. Cluster 2, which covers the medium production category, includes six sub-districts, namely Bilah Hulu, Panai Hulu, Panai Tengah, Rantau Selatan, and Rantau Utara. The resulting clustering is considered effective, with a silhouette score of 0.65609, which indicates that the data in the cluster is quite tight and clearly separated from the other clusters. This score indicates that the clustering was done well, resulting in a homogeneous and significant groupings. These findings are expected to contribute to the strategic planning and management of oil palm plantations in Labuhanbatu Regency, assisting the government in maximizing the potential of oil palm agricultural yields in the region. | en_US |
