<![CDATA[Apotek Naza plays an important role in providing medicines to the community. This study utilizes sales data from Apotek Naza for the period of July to December 2023. The K-Means algorithm is used to cluster the medicine data into clusters representing different sales patterns. The Elbow Method is employed to determine the optimal number of clusters (K) based on the Sum of Square Error (SSE). Evaluation is conducted using the Silhouette Coefficient (SC) to measure the quality of the resulting clusters. The analysis results show that the distribution of medicines in each cluster is as follows: 13.7% or 70 items are classified in the high-usage cluster (Cluster 0 - High), 57.5% or 294 items are classified in the medium-usage cluster (Cluster 1 - Medium), and 28.8% or 147 items are classified in the low-usage cluster (Cluster 2 - Low). This indicates a dominance of medium-usage medicines in the Apotek Naza dataset. The obtained Silhouette Score is 0.520, indicating that the clustering is well performed. According to Table 2.1 on the criteria for measuring clustering based on the Silhouette Coefficient (SC), this score indicates that the resulting clusters are fairly compact and well-separated from each other. Keywords: Medicine Inventory, Data Mining, K-Means, KDD, Elbow Method, Silhouette Coefficient]]>
Copyrights © 2024
