<![CDATA[This study applies the K-Means Clustering method to group sales transaction data from Cendana Shooping Online Store, located in Kefamenanu, North Central Timor Regency. The store utilizes social media as a means of promotion, communication, and transactions between sellers and customers. The analyzed dataset consists of 1,035 sales transactions recorded from January 2024 to March 2025, using item price, purchase quantity, and total payment as the clustering variables. The optimal number of clusters was determined using the Elbow Method, resulting in four clusters (k = 4). The clustering process converged after seven iterations, indicating that stable cluster centroids had been reached. Sales performance categories were determined based on the ranking of the average Total_Payment values of the final centroids. Cluster 3 was categorized as the very high-sales group, with an average Total_Payment of IDR 163,333.30 and 90 transactions; Cluster 2 as the high-sales group, with an average of IDR 119,098.40 and 61 transactions; Cluster 1 as the moderately high-sales group, with an average of IDR 77,817.57 and 296 transactions; and Cluster 4 as the low-sales group, with an average of IDR 41,938.78 and 588 transactions. Evaluation using the Silhouette Coefficient yielded an average score of 0.6246, indicating an acceptable clustering quality. The findings can serve as a basis for inventory management, marketing strategy development, and business decision-making.]]>
Copyrights © 2026
