Predicting Student Academic Performance by Mining Their Internet Usage Behaviour

DOI

10.2991/icedutech-17.2018.43How to use a DOI?

Keywords

student; academic performance; internet; behavior; naive bayes;

Abstract

Students who are active this time belong to the Generation Z. Namely, the generation that most of his time is spent interacting with computing devices, especially the virtual world. It can bring positive and negative effects on their performance in the learning process. This study proposes the use of students' internet behavior data to predict their academic performance. The method used in this study is Naive Bayes. There are six attributes of student internet behavior used in this study, including the number of social media accounts owned, the number of hours spent accessing social media, the number of hours spent on non-social media entertainment on the internet, the number of hours on the internet used for learning, and number of internet sessions in a week. The accuracy and sensitivity metric values of the experiments exceeded 85%, so it can be concluded that the mining of students' internet behavior data has the potential to be used to assist educational institutions in monitoring the performance of their students.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Utomo Pujianto
AU  - Mahmuda Muthmainnah
PY  - 2017/11
DA  - 2017/11
TI  - Predicting Student Academic Performance by Mining Their Internet Usage Behaviour
BT  - Proceedings of the 2017 International Conference on Education and Technology (2017 ICEduTech)
PB  - Atlantis Press
SP  - 219
EP  - 221
SN  - 1951-6851
UR  - https://doi.org/10.2991/icedutech-17.2018.43
DO  - 10.2991/icedutech-17.2018.43
ID  - Pujianto2017/11
ER  -