Prediction of National Examination Question using C4.5 Algorithm
Nur Lukman (a*), Beki Subaeki (b), Hilman Nur Abdullah (c), Aldy Rialdy Atmadja (d), Mohamad Wildanuddin (e)
a,c,d,e. Department of Informatics, UIN Sunan Gunung Djati Bandung,
b. Department of Information, System Sangga Buana YPKP University Bandung
Abstract
The National Examination in 2018 is no longer the only determinant of student graduation from their school, however the score of the national examination in 2018 has decreased from average value of 55.51 to 52.96. One of the reasons is that students found difficulties to predict the questions that probably appeared on the exam sheet. This research aimed to help students predict the questions that probably appeared, so that the average value hopefully will be increased. The prediction of the questions that probably appeared on the exam had been done by processing data of the questions used for national examinations several years ago using C 4.5 Algorithm. C 4.5 Algorithm is a Data Mining technique that can classify predictive questions and it is able to form a prediction of questions on the exam. The result showed that the prediction of questions that probably appeared on exams by using Algorithm C 4.5 had good performance and its accuracy is 70 %
Keywords: National Examination; Data Mining; C 4.5 Algorithm
Topic: Computer Science