Understanding misinformation on Twitter in the context of controversial issues

Social media is slowly supplementing, or even replacing, traditional media outlets such as television, newspapers, and radio. However, social media presents some drawbacks when it comes to circulating information. These drawbacks include spreading false information, rumors, and fake news. At least three main factors create these drawbacks: The filter bubble effect, misinformation, and information overload. These factors make gathering accurate and credible information online very challenging, which in turn may affect public trust in online information. These issues are even more challenging when the issue under discussion is a controversial topic. In this thesis, four main controversial topics are studied, each of which comes from a different domain. This variation of domains can give a broad view of how misinformation is manifested in social media, and how it is manifested differently in different domains. This thesis aims to understand misinformation in the context of controversial issue discussions. This can be done through understanding how misinformation is manifested in social media as well as by understanding people’s opinions towards these controversial issues. In this thesis, three different aspects of a tweet are studied. These aspects are 1) the user sharing the information, 2) the information source shared, and 3) whether specific linguistic cues can help in assessing the credibility of information on social media. Finally, the web application tool TweetChecker is used to allow online users to have a more in-depth understanding of the discussions about five different controversial health issues. The results and recommendations of this study can be used to build solutions for the problem of trustworthiness of user-generated content on different social media platforms, especially for controversial issues.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Informatics

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2019

Author dc:creator

• Addawood, Aseel

Contributors dc:contributor

• Ravaioli, Umberto
• Tewksbury, David
• Sobh, Nahil
• Stern, Chadly

• Social Media Text Analysis

Statement dc:rights

• Copyright 2019 Aseel Addawood

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/104800

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/104800