Aplikasi Data Mining untuk Prediksi Tingkat Kelulusan Siswa Menggunakan Metode SMOTE dan K-NN pada SMK PAB Helvetia

Abstract

Student graduation rate is the main indicator of the success of an educational institution, including SMK PAB Helvetia. However, the inability of schools to accurately predict graduation often becomes an obstacle in planning learning strategies. This study aims to develop an application for predicting student graduation rates using the Synthetic Minority Oversampling Technique (SMOTE) and K-Nearest Neighbor (K-NN) methods. SMOTE is used to handle student data imbalance, while K-NN is applied as a classification algorithm to predict graduation. Student data consisting of academic grades, attendance, and other parameters are processed through the designed application. The results of the study show that the combination of the SMOTE and K-NN methods can increase the accuracy of graduation predictions to achieve significant results compared to conventional methods. This application is expected to help schools identify students with low graduation risk and provide appropriate intervention recommendations. In addition, the results of this study can be a reference for further development in the application of data mining to support decision making in the education sector