Uncovering Cyberbullying Themes from Unconsented Facebook Pitik Post Through Text Mining Techniques

Abstract

Pitik, referred as street photography; is a colloquial phrase for taking random photographs, arbitrarily from people without permission and/or consent. Empirically, numerous comments circulating in Facebook Pitik posts containing words related to humiliation, embarrassment, shaming, stalking, and bullying. There are yet no studies conducted to confirm or prove the existence and extent of cyberbullying themes in Facebook Pitik trends. Cyberbullying is not new to social media environments and as technologies and trends change over time, the medium of cyberbullying also changes. We show in this study the proofs of the existence and the extents of cyber bullying themes in Facebook Pitik posts. In this study, we utilized methods of natural language processing – specifically text mining and emotion polarity computation, also known as sentiment analysis. Using Facepager software, 68,000 documents/comments are collected from select Facebook pages of photographers involved in the trend of Pitik posting. Results showed that the collected documents contain 26.29 % pertaining to harassment; 35.48% to flaming; and 19.45% to denigration. The existence of negative emotions is also seen from collected documents including anger, uncertainty, constraining, fear, sadness, and disgust. Findings may help policy makers to enhance the Facebook community standards making its app safer and free from issues relating to cyberbullying, especially in unconsented Pitik posts.