Klaster Kelompok Belajar Siswa Menggunakan K-Means dan Visualisasi Dashboard Power BI di SMPN 9 Mandau

At SMPN 9 Mandau, students are typically grouped based on random criteria, resulting in significant disparities in academic ability among study groups. Such an approach constrains teachers’ capacity to design effective instructional strategies that accommodate students’ learning preferences and needs. Alternatively, forming study groups using learning outcome data allows for clustering students with similar characteristics, thereby supporting more appropriate, student-centered grouping decisions. This study utilizes grade 9a student report card data from SMPN 9 Mandau. The analytical process integrates Python, SQL Server 2019 Management Studio, and Microsoft Power BI. The K-Means clustering algorithm is applied, resulting in four study groups: Cluster 1 consisting of eleven students, Cluster 2 of five students, Cluster 3 of four students, and Cluster 4 of eleven students. A multi-metric approach, including inertia with the elbow method, Silhouette Score, and Davies Bouldin Index, is used to determine the optimal number of clusters. The effectiveness of the developed business intelligence dashboard is further assessed through a user acceptance test, which yields a satisfaction score of 91.3%.