Newsletter
COSMOS Wins “Best Paper Award” for Groundbreaking AI-based Media Innovation Study at International AIMEDIA 2026 Conference in Nice, France
Our research titled “Symbolic Communication in the 2025 Tariff Discourse: A Comparative Analysis of X and Weibo” received the Best Paper Award at the International Conference on AI-based Media Innovation (AIMEDIA 2026), held July 5 to 9, 2026, in Nice, France. The award-winning study compares the narrative dynamics of trade-war discourse between Chinese social media and Western platforms, while identifying the drivers of discourse in terms of social, cultural, economic, and political symbols.
COSMOS Takes Center Stage at the 32nd AMCIS 2026 Conference in Reno, NV, USA
COSMOS made a splash at the 32nd Americas Conference on Information Systems (AMCIS 2026) in Reno, Nevada. Organized by the Association for Information Systems, AMCIS brings together leading researchers to address critical socio-technical challenges in digital platforms. Supported by federal grants, our studies examine how community expertise counters disguised toxicity, how algorithmic layout and persuasive messaging create content traps on YouTube, and how multimodal cultural symbols shape public trust during high-stakes elections. By uniting network science, dual-mechanism cognitive theory, and advanced AI vision-language models, this body of work delivers auditable, real-world solutions for modern platform governance.
Hot off the Press: Decoding Cross-Platform Tariff Discourse with W-CFSA
COSMOS recently published a new study in Springer’s Journal of Social Network Analysis and Mining that introduces Weighted Contextual Focal Structures Analysis (W-CFSA), an advanced graph-based framework designed to identify cohesive narrative formations across multiplex networks. While traditional social network analysis often evaluates platforms in isolation or relies on unweighted structural ties, W-CFSA integrates weighted interaction strengths, semantic narrative similarities, and cross-platform user dynamics. Tested on over 70,000 posts across YouTube, TikTok, X, and Instagram regarding U.S.-China tariff policies, W-CFSA successfully uncovered tightly knit, cross-platform narrative blocs. The method demonstrated significantly higher internal cohesion and global network impact compared to standard eigenvector-centrality baselines.
Research Spotlight: How Grievances, Narratives, and Visual Symbols Spread Online
We spotlight three COSMOS studies presented at the 14th International Conference on Complex Networks and their Applications in New York, USA, that investigate how policy shocks, visual symbols, and economic disputes spread through digital populations. By combining advanced language models, computer vision frame extraction, and classical disease propagation models, the research tracks how individual posts coalesce into organized grievance communities and viral narrative outbreaks. Together, these studies introduce rigorous socio-computational methods for auditing online influence, predicting the pre-mobilization phases of collective action, and mapping how multimodal content travels through complex networks.
New Hire: Vida Torgbe hired as Graduate Assistant at COSMOS Research Center
COSMOS welcomes Vida Torgbe as a new Graduate Assistant, where she contributes to research on social media behaviors and dissemination efforts. She is currently pursuing an MS in Information Science at UA Little Rock and brings a multidisciplinary background in social sciences and data analysis. At COSMOS, she works with Prof. Nitin Agarwal to strengthen her skills in social media analytics, data science, and interdisciplinary research while contributing to projects that address complex societal challenges.