Implementation K-Means Algorithm and Decision Tree for Prediction Graduation in Vocational School Student Diploma Data

Yohanes Brekmans M Darkel (Universitas Nusa Nipa, Indonesia)
Hilarius Alfan (Universitas Nusa Nipa, Indonesia)
Gerfasius Take Piran (Universitas Nusa Nipa, Indonesia)
Henderikus Basilius Nong Muda (Universitas Nusa Nipa, Indonesia)

G-Tech : Jurnal Teknologi Terapan
Vol 10 No 3 (2026): G-Tech, Vol. 10 No. 3 July 2026

This study aims to analyze vocational high school (SMK) students' diploma data using two data mining approaches: the K-Means algorithm for clustering and the Decision Tree for classification and graduation prediction. The application of these two methods provides insights into the clustering structure of students based on their diploma scores and produces a more accurate graduation prediction model. Data processing was conducted using RapidMiner through preprocessing, K-Means clustering, and classification modeling with Decision Tree. The results indicate that K-Means successfully forms student clusters based on the similarity of their scores, while Decision Tree generates classification rules that can be used to predict graduation categories. This study contributes to academic evaluation and enhances data-driven decision-making in the school environment.