Analisis Klaster Siswa Unggulan dengan Algoritma K-Means Berdasarkan Aspek Akademik dan Non-Akademik

Abstract

Improving the quality of education does not only depend on academic aspects, but also needs to consider non-academic aspects such as extracurricular participation, social attitudes, and student attendance. Therefore, an analysis method is needed that is able to group students comprehensively based on these various indicators. This study aims to group class X students of SMAS Sultan Agung Puger based on academic and non-academic aspects using the K-Means Clustering algorithm. The data used include academic grades, involvement in extracurricular activities, achievement, social attitude values, and the number of absences. The data processing process is carried out through the pre-processing stage, data transformation, application of the K-Means algorithm, and evaluation of clustering results using the Davies-Bouldin Index (DBI) method. The results of the analysis show that the formation of 4 clusters is the optimal structure with a DBI value of 1.1601. Each cluster has different characteristics that reflect the level of student achievement in various aspects. This clustering provides useful information for schools in designing student development strategies based on group needs. These findings support the application of a data-based approach in decision making in the field of education.