Social media platforms like Twitter have changed the way we communicate, but they have also created new challenges regarding online safety and content quality. One key concern is the existence of fake accounts, which are frequently used to distribute harmful content, such as hate speech. These fake accounts reduce the reliability of online interactions and negatively impact the overall quality of discussions.
This thesis aims to address this issue by using the PageRank algorithm, a popular method originally developed by Google to rank web pages, to help identify fake accounts on social media networks based on user interactions. Following that, hate speech detection techniques will be used to analyze the posts made by these accounts. The main goal is to determine how much hate speech comes from fake accounts and understand how they contribute to the spread of harmful content online.
Supervisor: Mahboobeh Zangiabady