In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
Niloofar Safi Samghabadi
will defend her proposal
Automatic Detection of Nastiness and Early Signs of Cyberbullying Events on Social Media
AbstractIn this thesis, we propose a Natural Language Processing system to predict the early signs of cyberbullying on social media. Our system will continuously monitor online content, provide timely predictions, and trigger an alert where the risk of cyberbullying is high. Although social media has made it easy for people to connect on a virtually unlimited basis, it has also opened doors to people who misuse it to undermine, harass, humiliate, threaten and bully others. Cyberbullying is a serious issue in cyberspace. Due to the growing number of online users, manual moderation of online content is impractical. Available automatic systems for hate speech and cyberbullying detection fail to do on-line predictions, which makes them ineffective for prevention. In this thesis, we will present new data resources as well as new tools for detecting aggression and early signs of cyberbullying in social media. First, we develop computational methods to identify aggressive content by characterizing online posts. We will further expand these approaches to model the dynamics of users' interactions on social media. Ultimately, we plan to design a system that monitors the threads of messages for online users and detects early signs of cyberbullying accurately and based on limited evidence.
Date: Monday, November 25, 2019
Time: 10:00 AM - 12:00 PM
Place: PGH 550
Advisor: Dr. Thamar Solorio
Faculty, students, and the general public are invited.