Multiscale online media simulation with SocialCube

Abdelzaher, Tarek; Hao, Yifan; Jing, Andong; Liu, Dongxin; Liu, Shengzhong; Nguyen, Hoang Hai; Nicol, David M.; Shao, Huajie; Wang, Tianshi; Yao, Shuochao; Malik, Omar; Dipple, Stephen; Flamino, James; Buchanan, Fred; Cohen, S.; Korniss, G.; Szymański, Bolesław K.

doi:10.1007/s10588-019-09303-7

Cited by 12 publications

(5 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By randomly assigning participants to engage in various online environments where the content, features, and users within the environment have been purposefully manipulated by the research team, our platform provides us with the ability to experimentally isolate the effects of each on participants' attitudes and online behaviors. 8 Simulated social media environments are emerging as useful tools for examining the impacts of social media, especially in the field of computer science (Abdelzaher et al 2020;Mahajan et al 2021;Masur, DiFranzo, and Bazarova 2021), though ours is the first that we are aware of that incorporate large language models to interact dynamically and directly with respondents in real-time.…”

Section: Methodsmentioning

confidence: 99%

Outnumbered Online: The Consequences of Partisan Imbalance in Online Political Discussions

Allamong,

Trexler,

Alqabandi

et al. 2024

Preprint

View full text Add to dashboard Cite

Research on the impacts of online political discussions have focused on social media "echo chambers," but less is known about how people respond to online environments dominated by those who are politically dissimilar. We conduct a preregistered experiment using a mobile application we developed to evaluate how being outnumbered by out-partisans impacts comfort with sharing opinions as well as perceptions of the platform and its users. Our app mimics a social media platform but provides researcher control over platform features to experimentally isolate their effects and uses automated chatbots to create a dynamic newsfeed for the participant. We find that engaging on a platform dominated by out-partisans reduces comfort with sharing one's opinions and lowers evaluations of the platform and its users. These findings shed light on a relatively less explored online environment (i.e., an outnumbered setting) and highlight the utility of LLMs in social science research.

show abstract

Section: Methodsmentioning

confidence: 99%

Outnumbered Online: The Consequences of Partisan Imbalance in Online Political Discussions

Allamong,

Trexler,

Alqabandi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The underlying assumption of this model is that the immediate future of the time series will simply replicate its immediate past. Within the realm of social media time series prediction, this baseline has been used in [17], [18], [34].…”

Section: B Baselines Usedmentioning

confidence: 99%

Simulating User-Level Twitter Activity with XGBoost and Probabilistic Hybrid Models

Mubang¹,

Hall²

2022

Preprint

View full text Add to dashboard Cite

The Volume-Audience-Match simulator, or VAM was applied to predict future activity on Twitter related to international economic affairs. VAM was applied to do timeseries forecasting to predict the: (1) number of total activities, (2) number of active old users, and (3) number of newly active users over the span of 24 hours from the start time of prediction. VAM then used these volume predictions to perform user link predictions. A user-user edge was assigned to each of the activities in the 24 future timesteps. VAM considerably outperformed a set of baseline models in both the time series and user-assignment tasks.

show abstract

“…Agent-based-modeling (ABM) techniques use both statistical and machine learning regression methods to forecast individual user activity streams. Abdelzaher et al [14] represent each user's activity by a timeseries of K elements, where each element represents the user activity in an arbitrary time granularity (e.g., hours, days, etc.). They used both ARIMA and deep neural networks (such as CNN and RNN) to predict the next K elements of the timeseries.…”

Section: Related Workmentioning

confidence: 99%

Social-Media Activity Forecasting with Exogenous Information Signals

Ng,

Horawalavithana,

Iamnitchi

2021

Preprint

View full text Add to dashboard Cite

Due to their widespread adoption, social media platforms present an ideal environment for studying and understanding social behavior, especially on information spread. Modeling social media activity has numerous practical implications such as supporting efforts to analyze strategic information operations, designing intervention techniques to mitigate disinformation, or delivering critical information during disaster relief operations.In this paper we propose a modeling technique that forecasts topic-specific daily volume of social media activities by using both exogenous signals, such as news or armed conflicts records, and endogenous data from the social media platform we model. Empirical evaluations with real datasets from two different platforms and two different contexts each composed of multiple interrelated topics demonstrate the effectiveness of our solution.

show abstract

Multiscale online media simulation with SocialCube

Cited by 12 publications

References 27 publications

Outnumbered Online: The Consequences of Partisan Imbalance in Online Political Discussions

Outnumbered Online: The Consequences of Partisan Imbalance in Online Political Discussions

Simulating User-Level Twitter Activity with XGBoost and Probabilistic Hybrid Models

Social-Media Activity Forecasting with Exogenous Information Signals

Contact Info

Product

Resources

About