Abstract:Understanding political polarization on social platforms is important as public opinions may become increasingly extreme when they are circulated in homogeneous communities, thus potentially causing damage in the real world. Automatically detecting the political ideology of social media users can help better understand political polarization. However, it is challenging due to the scarcity of ideology labels, complexity of multimodal contents, and cost of time-consuming data collection process. In this study, we adopt a heterogeneous graph neural network to jointly model user characteristics, multimodal post contents as well as user-item relations in a bipartite graph to learn a comprehensive and effective user embedding without requiring ideology labels. We apply our framework to online discussions about economy and public health topics. The learned embeddings are then used to detect political ideology and understand political polarization. Our framework outperforms the unimodal, early/late fusion baselines, and homogeneous GNN frameworks by a margin of at least 9% absolute gain in the area under the receiver operating characteristic on two social media datasets. More importantly, our work does not require a time-consuming data collection process, which allows faster detection and in turn allows the policy makers to conduct analysis and design policies in time to respond to crises. We also show that our framework learns meaningful user embeddings and can help better understand political polarization. Notable differences in user descriptions, topics, images, and levels of retweet/quote activities are observed. Our framework for decoding user-content interaction shows wide applicability in understanding political polarization. Furthermore, it can be extended to user-item bipartite information networks for other applications such as content and product recommendation.

Heterogeneity of Facebook friend network facilitates political learning: Evidence from a panel survey during the 2016 US presidential campaign

Political science. Exposure to ideologically diverse news and opinion on Facebook

Like-minded sources on Facebook are prevalent but not polarizing

The knowledge gap on social media: Examining roles of engagement and networks

From observation on social media to offline political participation: The social media affordances approach

More than 'Left and Right': Revealing Multilevel Online Political Selective Exposure

Examining (Political) Content Consumption on Facebook Through Data Donation

How do social media feed algorithms affect attitudes and behavior in an election campaign?

Incidental News Exposure on Social Media: A Campaign Communication Mediation Approach

No Echo in the Chambers of Political Interactions on Reddit

Social Media for Political Campaigns: An Examination of Trump’s and Clinton’s Frame Building and Its Effect on Audience Engagement

Incidental Exposure and News Engagement: Testing Temporal Order and the Role of Political Interest

Understanding Political Polarization via Jointly Modeling Users, Connections and Multimodal Contents on Heterogeneous Graphs

Friendship Networks and Political Opinions: A Natural Experiment among Future French Politicians

Facebook, trust and privacy in an election year: Balancing politics and advertising

Network Strategies in Election Campaigns

Engaging with Political and Social Issues on Facebook in College Life

Keep Your Friends Close and Your Facebook Friends Closer: A Multiplex Network Approach to the Analysis of Offline and Online Social Ties

Facebook as an instrument of election campaigning and voters’ engagement: Comparing Czechia and Poland

Learning Political News From Social Media: Network Media Logic and Current Affairs News Learning in a High-Choice Media Environment

Keep Them Engaged! Investigating the Effects of Self-centered Social Media Communication Style on User Engagement in 12 European Countries