Klasterisasi Penempatan Siswa yang Optimal untuk Meningkatkan Nilai Rata-Rata Kelas Menggunakan K-Means

The implementation of learning by teachers can measure the quality of schools and students. Schools with diverse student backgrounds need to take strategic steps in managing learning to get optimal learning outcomes. Good learning designs and techniques can motivate students' interest in learning. The teacher's role is very important in managing learning to create an effective teaching and learning process. Data Mining or also known as Knowledge Discovery in Database (KDD) is the process of extracting knowledge from large data to find new patterns to get new knowledge and information. Data Mining technology is used to explore existing knowledge in the database. One of the methods used in data mining is clustering with the K-Means algorithm. This study aims to conduct student clustering to obtain a balanced class composition in order to improve the quality and student learning outcomes as seen in the increasing in the class average score. The data processed in this study came from the main school data as many as 90 students of the XI class of Computer Network Engineering Skills Competency at SMKN Negeri 2 Padang Panjang in the 2020/2021 school year. The variables used in data processing are student scores, parents' income and the distance from where students live to school. The student clustering calculation using K-Means succeeded in grouping 90 students into 3 clusters where cluster 1 totaled 47 students, cluster 2 totaled 10 students and cluster 3 totaled 33 students. Each member of the cluster will be divided evenly into 3 groups studying to get a balanced class composition. This research can be used as a basis for decision making by schools in clustering student placements to improve learning outcomes. By the increasing in the grade point average, the school average score will also increased.