PREDIKSI TINGKAT KELULUSAN MAHASISWA DENGAN ALGORITMA C4.5

Abstract

This study aims to predict the number of students who will graduate on time using 6 variables, namely cumulative grade point average (GPA), high school origin, active organization, active student activity unit (UKM), already working, and motivation. At the Modeling stage, 32 rules were produced which will be used as the basis for decisions for student graduation predictions and testing accuracy results from the graduation prediction process consisting of 64 test data and 600 training data, where testing data taken from 10% of the training data obtained accuracy results of 82.81% by testing using a confusion matrix, this means that the C4.5 algorithm can predict well whether students will graduate on time or not