Penerapan Algoritma K-Means dan Decision Tree Dalam Analisis Prestasi Siswa Sekolah Menengah Kejuruan

Abstract

This study aims to determine how much influence parents' income and the distance from a student's home to school have on student achievement at SMK Tarbiyatul Ulum. Because if you know the factors that influence student achievement, steps or actions can be taken to improve student achievement. The method used in this study includes the use of clustering using K-Means and then classifying it using the Decision Tree method with a total of 157 datasets used as research material. After conducting the classification modeling using the decision tree algorithm, it was found that parents' income did not affect student achievement, but the distance from home to school did affect student achievement. Then at the evaluation stage, the decision tree algorithm is not suitable for use in predicting student achievement, because the accuracy and AUC values of this algorithm are 68% and 0.561, where these values fall into the failure category.