Analyzing Political Polarization in Twitter
Event details
| Date | 03.09.2018 |
| Hour | 10:00 › 12:00 |
| Speaker | Elmas Tugrulcan |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam (retake)
Exam president: Prof. Robert West
Thesis advisor: Prof. Karl Aberer
Co-examiner: Prof. Pierre Dillenbourg
Abstract
The cognitive dissonance phenomena states that people feel a mental discomfort when they are exposed to information that challenges their prior beliefs. This causes homophily which is socializing with people of same beliefs, and selective exposure, which is choosing information sources that confirms one's prior beliefs. Those in turn render people grow too confident of their political beliefs and take their political attitudes to extreme, which leads to phenomena called "Polarization". Today there is a raising concern that Twitter may lead to more polarization due to its features that enhances such homophily and selective exposure. There is an ongoing research to assess the polarization in Twitter and how to reduce it.
In this report, we will first analyze a classic paper that demonstrates the polarization in Twitter. We will then proceed by a work that asses polarity of users and news based on social ties and offers solution to reduce polarization. Lastly, we will analyze a paper that argues the news that are previously assumed to be polarized have very small absolute difference in bias except for special cases. In our research propose we will propose to extend assessment of polarity by taking these cases into account and argue how could we use these cases to reduce polarization by decreasing information overload.
Background papers
Political Polarization on Twitter, by Conover et al.
Joint Non-negative Matrix Factorization for Learning Ideological Leaning on Twitter, by Lahoti et al.
Fair and Balanced? Quantifying Media Bias through Crowdsourced Content Analysis, by Budak et al.
Exam president: Prof. Robert West
Thesis advisor: Prof. Karl Aberer
Co-examiner: Prof. Pierre Dillenbourg
Abstract
The cognitive dissonance phenomena states that people feel a mental discomfort when they are exposed to information that challenges their prior beliefs. This causes homophily which is socializing with people of same beliefs, and selective exposure, which is choosing information sources that confirms one's prior beliefs. Those in turn render people grow too confident of their political beliefs and take their political attitudes to extreme, which leads to phenomena called "Polarization". Today there is a raising concern that Twitter may lead to more polarization due to its features that enhances such homophily and selective exposure. There is an ongoing research to assess the polarization in Twitter and how to reduce it.
In this report, we will first analyze a classic paper that demonstrates the polarization in Twitter. We will then proceed by a work that asses polarity of users and news based on social ties and offers solution to reduce polarization. Lastly, we will analyze a paper that argues the news that are previously assumed to be polarized have very small absolute difference in bias except for special cases. In our research propose we will propose to extend assessment of polarity by taking these cases into account and argue how could we use these cases to reduce polarization by decreasing information overload.
Background papers
Political Polarization on Twitter, by Conover et al.
Joint Non-negative Matrix Factorization for Learning Ideological Leaning on Twitter, by Lahoti et al.
Fair and Balanced? Quantifying Media Bias through Crowdsourced Content Analysis, by Budak et al.
Practical information
- General public
- Free
Contact
- EDIC - [email protected]