Publication Date

Fall 2021

Degree Type

Master's Project

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Fabio Di Troia

Second Advisor

Navrati Saxena

Third Advisor

Mahesh Jaliminche

Keywords

Benford’s Law, Twitter, Machine Learning, Social Bots

Abstract

Over time Online Social Networks (OSNs) have grown exponentially in terms of active users and have now become an influential factor in the formation of public opinions. Due to this, the use of bots and botnets for spreading misinformation on OSNs has become a widespread concern. The biggest example of this was during the 2016 American Presidential Elections, where Russian bots on Twitter pumped out fake news to influence the election results.

Identifying bots and botnets on Twitter is not just based on visual analysis and can require complex statistical methods to score a profile based on multiple features and compute a result. Benford’s Law or the Law of Anomalous Numbers states that in any naturally occurring sequence of numbers, the first significant leading digit frequency follows a particular pattern such that they are unevenly distributed and reducing. This principle can be applied to the first- degree egocentric network of a Twitter profile to assess its conformity to Benford’s Law and classify it as a bot profile or normal profile.

This project focuses on leveraging Benford’s Law in combination with various Machine Learning (ML) classifiers to identify bot profiles on Twitter. In addition, the project also discusses various statistical methods that are used to verify the classification results.

Download

Share

COinS