Comparison of the Performance of the k-Nearest Neighbor, Naïve Bayes Classifier and Support Vector Machine Algorithm With SMOTE for Classification of Bully Behavior on the WhatsApp Messenger Application

DOI

10.2991/assehr.k.201230.028How to use a DOI?

Keywords

Cyberbullying, WhatsApp, k-NN, NBC, SVM, SMOTE

Abstract

WhatsApp is the most popular messaging application in Indonesia. This causes the emergence of cyberbullying behavior by its users. Cyberbullying is a dangerous problem because it has a very serious impact on the victim’s psyche such as feelings of hurt and disappointment. This study aims to classify WhatsApp chat into two classes, namely, bully and not bully. The classification algorithms used are k-NN, NBC, and SVM with SMOTE. The results show that the SVM algorithm with SMOTE is better at solving this case with an accuracy of 83,57%.

Copyright

© 2020, 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  - Irwansyah Saputra
AU  - Puput Irfansyah
AU  - Erlando Doni Sirait
AU  - Dwi Dani Apriyani
AU  - Michael Sonny
PY  - 2020
DA  - 2020/12/31
TI  - Comparison of the Performance of the k-Nearest Neighbor, Naïve Bayes Classifier and Support Vector Machine Algorithm With SMOTE for Classification of Bully Behavior on the WhatsApp Messenger Application
BT  - Proceedings of the 1st International Conference on Folklore, Language, Education and Exhibition (ICOFLEX 2019)
PB  - Atlantis Press
SP  - 143
EP  - 149
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.201230.028
DO  - 10.2991/assehr.k.201230.028
ID  - Saputra2020
ER  -