Klasifikasi Performa Akademik Siswa Menggunakan Metode Decision Tree dan Naive Bayes

Abstract

Getting good academic performance is the goal of the learning process carried out by the education office under the auspices of the Ministry of Education and supervised by the government. Governments that want to be successful in educating students should pay attention to their new generation because they are the future successors of the nation. Students of all levels are the benchmark for a country's success.  Therefore, it is necessary to know the student's academic performance from an early age, in order to get special treatment related to the student's learning achievement.  In this study, the academic achievement of students from various levels of education such as elementary, middle, and high schools was tried to be determined by applying various data mining classification methods such as Decision Tree and Naive Bayes. This data grouping is open where the data can be accessed easily and can be used in future research.  The data grouping is divided into 3 categories, namely Low (L), Medium (M) and High (H). From the results of the dataset trial, it shows that the highest classification accuracy is Decision Tree of 83.89% and Naive Bayes of 85.97%. Thus the Naïve Bayes method is more accurate in grouping students' academic performance data.   The results of this study will be used to create a student grouping system based on student learning performance using the appropriate algorithm. So that his contribution later when creating the application system for the formation of study groups can use the Naïve Bayes method.