The International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction and Behavior Representation in Modeling and Simulation (SBP-BRiMS) leads interdisciplinary research in social computing and behavior modeling. As the name implies, the conference focuses on social and behavioral dynamics, which are critical for understanding complex phenomena in fields such as public health, cybersecurity, and national defense, with roots in diverse communities including academia, government, and industry. This research—driven largely by analyzing data from social media, simulations, and other behavioral data sources—commonly uses computational modeling, machine learning, and simulation techniques. It reflects the growing importance of data-driven insights in addressing social and behavioral challenges across sectors.
SBP-BRiMS 2024 aimed to unite experts across social science, computational modeling, and data science to advance methods for understanding and predicting complex human behavior. The event focused on addressing critical challenges in social computing, including modeling social influence, simulating cultural dynamics, and predicting behavioral trends. It invited rigorous theoretical research alongside applied studies that leverage large-scale data, machine learning, and simulation tools, highlighting applications in critical areas such as epidemiology, defense strategy, cybersecurity, and crisis response.
From September 18 to 20, 2024, the 17th SBP-BRiMS conference was held at Pittsburgh, PA. This year COSMOS had 6 papers accepted and presented at the conference. Several cosmographers traveled to the conference with Prof. Agarwal to present their research with travel support from Prof. Agarwal. The following is a list of papers from cosmographers that were published in the conference:
- Analyzing and Predicting Meetup Mobs Outcome Via Statistical Analysis and Deep Learning
- Ablation Studies in Protest Networks: The Role of Influential Agents in Shaping Protests
- Unveiling Bias in YouTube Shorts: Analyzing Thumbnail Recommendations and Topic Dynamics
- Investigating YouTube Narratives and User Resonance in the South China Sea Dispute.
- Mobilization Characteristics of Disinformation vs Anti-Disinformation campaign on TikTok During Taiwan’s 2024 Presidential Elections
- Developing Epidemiological Models with Differentiated Infected Intensity
In this first article of a two-part series, we summarize three of these papers:
- Analyzing and Predicting Meetup Mobs Outcome Via Statistical Analysis and Deep Learning
- Ablation Studies in Protest Networks: The Role of Influential Agents in Shaping Protests
- Mobilization Characteristics of Disinformation vs Anti-Disinformation campaign on TikTok During Taiwan’s 2024 Presidential Elections
Each of these studies furthers collective action research. The latter two studies specifically focus on protests in times of election, while the first researches flash mobs that used Meetup.com.
“Analyzing and Predicting Meetup Mobs Outcome Via Statistical Analysis and Deep Learning” studied digital-era gatherings, or flash mobs, that are coordinated through social media, email, or texts. While these assemblies may appear random, they involve complex coordination. The study analyzed Meetup.com, an Event-Based Social Network, to understand group behavior patterns. Through statistical analysis and deep neural networks, the authors developed two predictive models that accurately forecasted whether these organized gatherings would succeed or fail.
“Ablation Studies in Protest Networks: The Role of Influential Agents in Shaping Protests” examined the role of influential social media agents and multimedia content in shaping protest movements, focusing on the 2022 Brazilian insurrection. The study analyzed Instagram posts using network analysis, emotion detection, and moral foundation theory. The findings showed that influential agents significantly amplified emotional responses, particularly anger, through both images and text during key protest phases. The research demonstrates how different content modalities contributed to collective identity formation and protest mobilization.
“Mobilization Characteristics of Disinformation vs Anti-Disinformation campaign on TikTok During Taiwan’s 2024 Presidential Elections” examined the mobilization characteristics of disinformation and anti-disinformation campaigns on TikTok during Taiwan’s 2024 presidential elections. The study used Diffusion of Innovations theory to analyze how information spreads through social networks and mobilizes users. Analyzing 343 TikTok videos and 46,551 comments, the researchers found that successful campaigns depended more on sustained audience engagement than content volume. While disinformation campaigns used rapid, high-volume bursts, anti-disinformation efforts achieved greater success through sustained, credible engagement.