<![CDATA[Social media is popular communication platform for last decade. Social Platform such as Facebook, Instagram, and Twitter provide real-time and efficient way of communication overseas. The ease of using social media does not only provide positive benefits, but can also have a negative impact on its users. One of social media negative impact is cyber-bullying which define as a type of harassment through online media. The effect of cyber-bullying to the victim particularly is mental health disorder. Usually, being the victim of cyber-bullying can increase the stress and anxiety level, lower self-esteem, loneliness, sadness, and disappointment. This study evaluates the comment on Instagram post of Indonesia influencer to determine whether it classified as bullying or non-bullying. This study utilizes count vectorizer as feature extraction and compare several machine learning methods such as Naïve Bayes, SVM, and Random Forest. The evaluation result show that both Naïve Bayes and Random Forest are able achieve 77% accuracy. Moreover, Naïve Bayes method also generate higher percentage compared to other methods. This result indicate that Naïve Bayes are capable in detecting cyber-bullying comment in social media platform.]]>
Copyrights © 2023
