Online interactions comprise a combination of positive and negative exchanges, shaping complex social dynamics. Signed network representations—where ties can be either positive or negative—provide a powerful lens for studying these dynamics, particularly in the context of political polarization. In this talk, I present an analysis of polarization on Menéame, a Spanish social media platform centered on news sharing, commenting, and voting. Using a dual-method approach—Signed Hamiltonian Eigenvector Embedding for signed networks and Correspondence Analysis for unsigned networks—we examine how antagonistic interactions enrich our understanding of structural polarization. Results show that while unsigned networks successfully map ideological communities, incorporating negative ties is essential to reveal ideologically extreme users who engage in antagonistic behaviors. Without this perspective, the most confrontational actors remain hidden among their less combative peers, underscoring the critical role of negative ties in mapping online polarization.
I have been actively involved in the committees organizing WWCS, WiNS Seminars, yrCSS, etc. These experiences have given me the opportunity to connect with amazing people, exchange ideas, and build strong bonds across the world. At the same time, I also recognize that such organization tasks are very time-consuming and often not formally acknowledged as part of the PhD journey. This brings up important questions: in what ways can institutions better acknowledge and value the contributions made through organizing academic events, and how can we, as PhD students, manage these responsibilities alongside our other academic commitments?
Can we find polarizing users in flame wars? In this seminar, Elena Candellone (Utrecht University), presented network embedding methods for online social interactions and opened a discussion on the benefits of organizing scientific events and initiatives.
Elena presented work on online ecosystems and their role in amplifying polarization. By analyzing a Spanish Reddit-like platform, she applied network embeddings to capture sentiment outliers, propaganda, and industry outlets. Her results suggest certain embedding methods are better suited for signed networks with negative interactions.
In the discussion, Elena also shared her experiences organizing scientific events. She highlighted the joy and personal rewards of connecting people, despite the challenges of balancing these tasks with academic commitments. Though such efforts are not always suffiecienty appreciated professionally, they are increasingly valued in grants and academic positions. The central questions raised were: how can institutions better acknowledge organizational work, and how can early-career researchers balance the triple role of research, teaching, and outreach?
