Claim Missing Document
Check
Articles

Found 1 Documents
Search

Aplikasi Data Mining menggunakan Algoritme C4.5 untuk Memprediksi Mahasiswa Berpotensi Drop Out Izza Isma; Ahmad Afif Supianto; Andi Reza Perdanakusuma
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 7 (2019): Juli 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (818.501 KB)

Abstract

Higher education is the organizer of academic education for students where the quality of a college can be seen from the high level of success of students and the low failure rate of students. One indicator of student failure is a drop out case where it is also experienced by the study program Information System of the Faculty of Computer Science, University of Brawijaya. Based on the results of the interview obtained information that every year there are students of Information Systems who resign. The case certainly needs to be considered so as not to reduce the quality of education and university accreditation. So based on these problems a system is needed that is able to be a decision support to detect students who have the potential to drop out so that further action can be given. C4.5 algorithm is one of the algorithms in data mining that can be used to predict students who have the potential to drop out by generating rule in form of decision tree. The attributes used consist of academic data and student demographics of the study program Information Systems. The results from the calculation of confussion matrix are an accuracy rate of 98.85%. Meanwhile, based on the ROC curve, the AUC value is 0.8462. The output of this study is to make predictive applications for students who have the potential to drop out for the study program Information System by applying the C4.5 algorithm for the mining process on a system developed in the form of a website-based dashboard. Usability testing using the System Usability Scale (SUS) is 67.5 which is included in the adjective rating Good category and the acceptability level of users is included in the Marginal-High category.