Agent-based simulation of online social networks and disinformation

2025-12-26

AI Summary

Online social networks (OSNs) suffer from platform opacity, restricted data access, and a lack of realism and interpretability in existing simulation approaches. Method: This paper introduces the first high-fidelity, interpretable OSN simulation framework integrating demographic-driven personality modeling, finite-state machine (FSM)-based behavioral automation, and large language model (LLM)-generated content synthesis. It operationalizes the DISARM countermisinformation workflow to model malicious information diffusion and implements visual validation via a Mastodon-inspired interface. Contribution/Results: The framework unifies structural realism (graph topology), behavioral interpretability (via FSMs), and semantic authenticity (via LLM-generated content). Experiments demonstrate that its topological metrics and generated content quality match those observed in real-world OSNs. The framework enables controllable, reproducible, and mechanistic studies of information diffusion, advancing both methodological rigor and explanatory power in computational social science.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Research on online social networks (OSNs) is often hindered by platform opacity, limited access to data, and ethical constraints. Simulation offer a valuable alternative, but existing frameworks frequently lack realism and explainability. This paper presents a simulation framework that models synthetic social networks with agents endowed with demographic-based personality traits and finite-state behavioral automata, enabling realistic and interpretable actions. A generative module powered by a large language model (LLM) produces context-aware social media posts consistent with each agent's profile and memory. In parallel, a red module implements DISARM-inspired workflows to orchestrate disinformation campaigns executed by malicious agents targeting simulated audiences. A Mastodon-based visualization layer supports real-time inspection and post-hoc validation of agent activity within a familiar interface. We evaluate the resulting synthetic social networks using topological metrics and LLM-based content assessments, demonstrating structural, behavioral, and linguistic realism. Overall, the framework enables the creation of customizable and controllable social network environments for studying information dynamics and the effects of disinformation.

Problem

• Simulates realistic online social networks with agent-based models
• Generates context-aware social media posts using large language models
• Models disinformation campaigns for studying information dynamics effects

Innovation

• Agent-based simulation with demographic traits and behavioral automata
• LLM-powered generative module for context-aware social media posts
• DISARM-inspired red module orchestrating disinformation campaigns