COMPARISON OF DECISION TREE AND NAÏVE BAYES ALGORITHMS IN PREDICTING STUDENT GRADUATION AT YPK JUNIOR HIGH SCHOOL, NABIRE REGENCY

Abstract

This study aims to compare the accuracy of the Decision Tree C4.5 and Naive Bayes algorithms in predicting student graduation at YPK Immanuel Nabire Junior High School, Central Papua. Student data from the 2022 and 2023 school years were used as training data, whereas student data for the 2024 school year were used as testing data. Data collection methods included field studies, interviews with schools, and literature studies. The implementation of the algorithm is carried out using the Orange software, which simplifies the process of data visualization and analysis. Both algorithms are applied to data processed through stages of cleaning and normalization to ensure the quality and relevance of the data used. The results show that the Decision Tree C4.5 algorithm has a prediction accuracy of 90.91%, while the Naive Bayes algorithm has an accuracy of 63.64%. The C4.5 Decision Tree algorithm is superior in predicting student graduation compared to Naive Bayes, which means that the C4.5 Decision Tree is more effective in identifying students who are likely to pass or not pass. The implementation of the C4.5 Decision Tree algorithm also helps schools make better decisions to support students who require additional attention. The study concluded that the Decision Tree C4.5 algorithm is recommended for use in predicting student graduation because it provides higher accuracy. The results of this research can be used by schools to improve the efficiency of the graduation prediction process and develop more effective and efficient learning programs. Using the right algorithms, schools can be more proactive in identifying students who need additional support, which can reduce academic failure rates and improve the overall quality of education