Implementation of Data Mining to Predict Student Graduation on Time using Random Forests

Abstract

The level of accuracy of student graduation in tertiary institutions is one of the criteria for assessing campus accreditation. The more students who graduate on time, the better the college's performance will be. Students' graduation rates are difficult to predict early, resulting in delays in graduation. To reduce the rate of delay in graduating college for students, it is necessary to be educated seriously in order to graduate on time. One method of solving this problem is by predicting the accuracy of student graduation by using data mining or data mining methods. The purpose of this system is to make it easier for lecturers on campus to classify students who are classified as graduating on time using the Random Forest method. The results of the classification using the Random Forest Algorithm using 1,351 data, then the evaluation results with an accuracy value of 90.74% by dividing the training and testing data as much as 80:20 The system successfully displays data visualization to predict graduation on time by implementing data mining.