Menentukan Prediksi Kelulusan Siswa Dengan Membandingkan Algoritma C4.5 Dan Naive Bayes Studi Kasus SMKN. 1 Cikarang Selatan

Abstract

The process of identifying information using statistical techniques, and machine learning is the meaning of data mining. Data mining can be applied in various fields of life such as health, business, and education. One of the applications of data mining in the field of education is to predict the graduation of school students. Prediction of student graduation using data derived from transcripts of the final grades of each student, while the attributes used are the average value of Indonesian, English, and Mathematics lessons from semester 1 to semester 5 as well as the history of SP that has been obtained during the student is in school. In this study, two data mining methods are used, namely the C4.5 algorithm and Naïve Bayes algorithm. The use of the two methods in this study aims to compare the performance of the two algorithms in predicting student graduation based on the level of accuracy, precision, and recall obtained. From the test results using data testing as much as 222 data which states that the C4.5 algorithm has an accuracy value of 98.64%, 100% precision and 100% recall, while Nave Bayes has an accuracy level of 97.75%, precision 95.52% and recall. 95.52%. And if the test uses 890 training data, it will state that the C4.5 algorithm has an accuracy level of 98.99%, precision 98.68% and recall 98.68% while nave Bayes has an accuracy level of 97.42%, precision 99, 39% and recalls 99.39%. From the above comparison, the C4.5 algorithm has an accuracy level that tends to be higher than the nave Bayes algorithm, so it was decided that in predicting student graduation, the C4.5 algorithm is better than the nave Bayes algorithm in predicting student graduation data.

Keywords: Data mining, Clacification, C4.5 Algoritma, Naive Bayes