Dipartimento di Informatica

Date: Fri 16 Oct 2020
Time: 08:00 AM
Venue: http://meet.google.com/xqj-mswe-wbv
Speaker: Walter Quattrociocchi

Title: From Confirmation Bias to Echo Chambers. A Data-driven approach.

Abstract: Do echo chambers actually exist on social media? By focusing on how both Italian and US Facebook users relate to two distinct narratives (involving conspiracy theories and science), we offer quantitative evidence that they do. The explanation involves users’ tendency to promote their favored narratives and hence to form polarized groups. Confirmation bias helps to account for users’ decisions about whether to spread content, thus creating informational cascades within identifiable communities. At the same time, aggregation of favored information within those communities reinforces selective exposure and group polarization. We provide empirical evidence that because they focus on their preferred narratives, users tend to assimilate only confirming claims and to ignore apparent refutations. We are working on a comparative study about the news consumption on Social Media on different European Countries (Germany, Spain, Italy and France).
It seems that Germany is the golden standard and example to follow to avoid polarization.
This perfectly fits on our hypothesis that the main driver behind misinformation is distrust in journalists and institution. Indeed, we are testing an automatic approach for the early detection of fake news topic by using as a proxy the polarizing effects of information on the public opinion (the accuracy for the detection is around 91%).
The detection of polarizing topics combined with the polarization rank (an indicator for each news outlet quantitatively describing the impact on the public opinion) would be useful tool to support journalist and policy maker to shape the communication strategies as well as to monitor the state of information ecosystem.