Detecting User Polarization in Social Media
Event details
| Date | 11.07.2018 |
| Hour | 14:00 › 16:00 |
| Speaker | Tugrulcan Elmas |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Robert West
Thesis advisor: Prof. Karl Aberer
Co-examiner: Prof. Pierre Dillenbourg
Abstract
Due to homophily principle, people only link themselves to likewise people, which leads to political polarization. Previous works indicate this phenomena is observed in Twitter users' retweets and hashtag usage. Polarization is detrimental, but it is helpful in predicting political orientation of users as users are coherent in their thoughts and behavior in Twitter.
We survey papers which exploit this phenomena by predicting political orientation by content and retweet behavior and propose our framework which will extend them by considering and taking advantage of all user features available.
Background papers
Political polarization on twitter, by M. D. Conover, et al.
Validation of Twitter opinion trends with national polling aggregates: Hillary Clinton vs Donald Trump, by Bovet, A., et al.
Biaswatch: A lightweight system for discovering and tracking topic-sensitive opinion bias in social media, by Lu, H., et al.
Exam president: Prof. Robert West
Thesis advisor: Prof. Karl Aberer
Co-examiner: Prof. Pierre Dillenbourg
Abstract
Due to homophily principle, people only link themselves to likewise people, which leads to political polarization. Previous works indicate this phenomena is observed in Twitter users' retweets and hashtag usage. Polarization is detrimental, but it is helpful in predicting political orientation of users as users are coherent in their thoughts and behavior in Twitter.
We survey papers which exploit this phenomena by predicting political orientation by content and retweet behavior and propose our framework which will extend them by considering and taking advantage of all user features available.
Background papers
Political polarization on twitter, by M. D. Conover, et al.
Validation of Twitter opinion trends with national polling aggregates: Hillary Clinton vs Donald Trump, by Bovet, A., et al.
Biaswatch: A lightweight system for discovering and tracking topic-sensitive opinion bias in social media, by Lu, H., et al.
Practical information
- General public
- Free
Contact
- EDIC - [email protected]