Defence of dissertation in the field of computer science, Kiran Garimella, M.Sc.

Measuring and Breaking Filter Bubbles on Social Media

Kiran Garimella, M.Sc., will defend the dissertation "Polarization on Social Media" on 7 February 2018 at 13 o'clock at the Aalto University School of Science.

Social media and the web have provided a foundation where users can easily access diverse information from around the world. Instead, many studies claim that social media has actually done the opposite — make people more close-minded and listen only to information they agree with. As an example, imagine two users of opposite ideological stances (liberal and conservative).  Though the two users may be looking at the same topic (e.g., a presidential debate), they might be seeing completely different viewpoints due to the diverse network surroundings and different content sources. Consequently, users on different ends of the ideological spectrum live in their own filter bubbles, oblivious to the views on the other side and creating their own world-view of truth. This phenomenon has led to the polarization of viewpoints, intolerance towards opposing views, and ideological segregation. This is a troubling phenomenon that can lead to societal divisions and hurt the democratic process.

In this thesis, we try to understand this phenomenon of polarization on social media.
We first start by building algorithms to detect polarized topics automatically from large social-media streams. Next, using these algorithms we study how polarization has evolved over time on Twitter. Finally, we propose methods to reduce polarization by connecting people to content outside their filter bubble. Given the impact of a polarized social media has on the society - from recent events such as Brexit and the U.S. elections - methods like the ones we propose in this thesis are the need of the hour. 

Dissertation release (pdf)

Opponent: Professor Dino Pedreschi, University of Pisa, Italiy

Custos: Professor Aristides Gionis, Aalto University School of Science, Department of Computer Science