Leveraging LLMs for Privacy-Aware Predictions in Participatory Budgeting

2025-08-07

AI Summary

This study addresses three key challenges in participatory budgeting (PB): heterogeneous proposal quality, low transparency in large-scale proposal review, and voter privacy leakage. To this end, we propose a privacy-preserving, large language model (LLM)-driven prediction framework. Methodologically, the framework leverages only proposal text and anonymized historical voting records—eschewing all demographic or personally identifiable information—and integrates GPT-4 Turbo’s in-context learning with behavioral modeling of past voting patterns. Crucially, it synergistically combines LLM-derived prior knowledge with de-identified collective decision-making patterns to achieve high-confidence funding predictions. Experimental results demonstrate that our framework significantly improves proposal optimization efficiency and review transparency, enabling scalable, auditable PB processes. Importantly, it achieves these gains while rigorously preserving voter privacy—thereby strengthening public trust and enhancing the legitimacy and accountability of democratic budgetary decisions.

Technology Category

• Machine Learning: Privacy
• Humans and AI: Voting
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Security and Privacy: Data transparency and provenance
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Participatory Budgeting (PB) empowers citizens to propose and vote on public investment projects. Yet, despite its democratic potential, PB initiatives often suffer from low participation rates, limiting their visibility and perceived legitimacy. In this work, we aim to strengthen PB elections in two key ways: by supporting project proposers in crafting better proposals, and by helping PB organizers manage large volumes of submissions in a transparent manner. We propose a privacy-preserving approach to predict which PB proposals are likely to be funded, using only their textual descriptions and anonymous historical voting records -- without relying on voter demographics or personally identifiable information. We evaluate the performance of GPT 4 Turbo in forecasting proposal outcomes across varying contextual scenarios, observing that the LLM's prior knowledge needs to be complemented by past voting data to obtain predictions reflecting real-world PB voting behavior. Our findings highlight the potential of AI-driven tools to support PB processes by improving transparency, planning efficiency, and civic engagement.

Problem

• Enhancing participatory budgeting with AI-driven proposal support
• Predicting funding outcomes using text and anonymous voting data
• Improving transparency and engagement in public budgeting processes

Innovation

• Leveraging LLMs for privacy-aware PB predictions
• Using textual descriptions and anonymous voting records
• Complementing LLM knowledge with historical voting data