Sentiment Analysis Of Cyberbullying On Twitter Using SentiStrength

Khaira, Ulfa and Johanda, Ragil and Utomo, Pradita Eko Prasetyo and suratno, tri (2020) Sentiment Analysis Of Cyberbullying On Twitter Using SentiStrength. Indonesian Journal of Artificial Intelligence and Data Mining (IJAIDM), 3 (1). pp. 21-27. ISSN 2614-6150

Text 5033 - Published Version Download (22kB)

Abstract

Cyberbullying is a form of bullying that takes place across virtually every social media platform. Twitter is a form of social media that allows users to exchange information. Bullying has been a growing problem on Twitter over the past few years. Sentiment analysis is done to identify the element of bullying in a tweet. Sentiments are divided into 3 classes, namely Bullying, Non-Bullying and neutral. There are three steps to classify cyberbullying i.e. collection of data set, preprocessing data, and classification process. This research used sentiStrength, an algorithm which uses a lexicon based approach. This SentiStrength lexicon contains the weight of its sentiment strength. The assessment results from 454 tweets data obtained 161 tweet non-bullying (35.4%), 87 tweet neutral (19.1%), and 206 tweet bullying (45.4%). This research produces an accuracy value of 60.5%.

Type:	Article
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Fakultas Sains dan Teknologi > Sistem Informasi
Depositing User:	Khaira
Date Deposited:	23 Mar 2021 07:57
Last Modified:	23 Mar 2021 07:57
URI:	https://repository.unja.ac.id/id/eprint/16970

Actions (login required)

View Item