In this month’s Research Spotlight, COSMOS highlights two studies presented at the 14th International Conference on Complex Networks and Their Applications (Complex Networks) in New York, USA. Led by Prof. Nitin Agarwal, the research examines two connected questions at the heart of modern digital influence: which groups drive information through social networks, and how narratives continue to spread once they cross from one platform to another?

The first study, “CI-FSA: Toward Scalable Discovery of Influential Groups in Social Networks,” challenges the idea that influence is driven primarily by a few powerful individuals. Instead, the research shows that influence often emerges from tightly connected groups working collectively. COSMOS introduces CI-FSA, a scalable method that identifies these influential subnetworks by considering both their structural cohesion and their ability to propagate information. Tested across multiple networks, including data from the 2024 Taiwan election and the Enron email communication network, the approach offers an interpretable way to uncover groups that may play important roles in coordinated influence, manipulation, and narrative formation.

The second study, “Modeling Cross-Platform Narrative Contagion Using a Multiplex SEIRZ Framework,” looks at what happens when narratives move beyond a single platform. Using data from Instagram, TikTok, X, and YouTube surrounding the 2024 Taiwanese presidential election, COSMOS models these platforms as interconnected layers of a larger information ecosystem. The findings show that narrative spread depends strongly on both exposure intensity and connections between platforms, demonstrating why studying social media platforms in isolation can overlook important pathways through which information gains reach and influence. 

Together, these studies provide a more complete picture of digital influence: connected groups can create the momentum behind information spread, while cross-platform pathways can extend that influence into new audiences and environments.

This work reflects Prof. Agarwal and COSMOS’s broader mission to develop scalable, interpretable methods for understanding complex online behavior. For science, these studies advance network analysis and cross-platform diffusion modeling. For society, they provide researchers, platforms, and policymakers with better tools to identify coordinated influence, monitor emerging narratives, and understand how information moves through increasingly interconnected digital ecosystems.