Comparative Analysis of Naive Bayes and C4.5 ALgorithms for Predicting Student Acceptance in Public Universities (Case Study: SMAN 2 Kota Bekasi)
Antonius Yadi Kuntoro (a*), Hermanto (b*), Hermanto Wahono (c*), Lasman Effendi (d*), Ridatu Oca Nitra (e*), Riza Pahlavi (f*), Mario Hengki (g*), Ferry Syukmana (h*), Sfenrianto (i*)
a) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*antonius.aio[at]nusamandiri.ac.id
b) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*hermanto.hmt[at]bsi.ac.id
c) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*mr.h3rm4n.gmail.com
d) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*lasman.lef[at]bsi.ac.id
e) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*ridatu.rdo[at]bsi.ac.id
f) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*riza.pahlavi[at]gmail.com
g) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*mariohengki[at]gmail.com
h) Master of Computer Science Postgraduate Program, STMIK Nusa Mandiri, Jakarta Indonesia
*14002191[at]nusamandiri.ac.id
i) Information System Management Department, BINUS Graduate Program - Master of Information Systems Management, Bina Nusantara University, Jakarta, Idonesia, 11480
*sfenrianto[at]binus.edu
Abstract
Learning achievement can be measured by school report cards. The challenges in this study can be done by using a classification method to predict learning achievement using classification algorithm namely Naive Bayes and C4.5. After a comparison between the two algorithm , the result of the prediction of learning achievement are obtained. It is clear that the naive bayes algorithm has an accuracy value of 69.18% and the AUC value of 0.771 is included in the fair classification, while the C4.5 algorithm has an accuracy of 65.65% and the AUC value of 0.686 is in poor classification. From these result it can be concluded that the naive bayes algorithm has a higher accuracy than the random forest algorithm and C4.5, the difference in accuracy between naive bayes and the difference between naive bayes and C4.5 is 3.53%. Thus the naive bayes algorithm can predict student achievement better
Keywords: C4.5, Naive Bayes, Student Achievement
Topic: International Symposium of Engineering, Technology, and Health Sciences