An AI-Powered Framework for Analyzing Collective Idea Evolution in Deliberative Assemblies

2025-09-15

AI Summary

Amid escalating social fragmentation, deepening political polarization, and declining institutional trust, this study addresses the core question in deliberative democracy: how individual opinions evolve into policy recommendations and subsequently influence voting behavior. Methodologically, it pioneers the systematic application of large language models (LLMs) to analyze transcribed deliberative meeting texts, introducing a framework for reconstructing individual opinion evolution trajectories. This framework integrates natural language processing with dynamic visualization to enable high-resolution modeling of the entire recommendation lifecycle—generation, convergence, and attrition. The contributions are threefold: (1) it overcomes limitations of conventional quantitative and qualitative approaches by precisely linking opinion shifts across multiple deliberation rounds to final voting outcomes; (2) it uncovers micro-level mechanisms underlying delegates’ positional reconfiguration during deliberation; and (3) it provides a reproducible, scalable, AI-augmented methodology for empirical research on deliberative democracy.

Technology Category

• Humans and AI: Voting
• Game Theory and Economic Paradigms: Social Choice / Voting
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

In an era of increasing societal fragmentation, political polarization, and erosion of public trust in institutions, representative deliberative assemblies are emerging as a promising democratic forum for developing effective policy outcomes on complex global issues. Despite theoretical attention, there remains limited empirical work that systematically traces how specific ideas evolve, are prioritized, or are discarded during deliberation to form policy recommendations. Addressing these gaps, this work poses two central questions: (1) How might we trace the evolution and distillation of ideas into concrete recommendations within deliberative assemblies? (2) How does the deliberative process shape delegate perspectives and influence voting dynamics over the course of the assembly? To address these questions, we develop LLM-based methodologies for empirically analyzing transcripts from a tech-enhanced in-person deliberative assembly. The framework identifies and visualizes the space of expressed suggestions. We also empirically reconstruct each delegate's evolving perspective throughout the assembly. Our methods contribute novel empirical insights into deliberative processes and demonstrate how LLMs can surface high-resolution dynamics otherwise invisible in traditional assembly outputs.

Problem

• Tracing idea evolution into policy recommendations
• Analyzing deliberative process impact on delegate perspectives
• Visualizing suggestion space and voting dynamics

Innovation

• LLM-based analysis of assembly transcripts
• Visualization framework for suggestion evolution
• Reconstruction of delegate perspective dynamics