PUBLICSPEAK: Hearing the Public with a Probabilistic Framework in Local Government

2025-03-14

AI Summary

To address the challenge of scalable identification of public statements in government open meetings, this paper proposes a probabilistic framework integrating structural priors and domain-specific semantic embeddings. Methodologically, it introduces the first joint modeling of meeting agenda structure, governmental domain knowledge, and linguistic features—implemented via a structure-aware encoder and a Bayesian graphical model—augmented with an interpretable uncertainty calibration mechanism to handle low-resource, high-noise civic text. Evaluated on a novel dataset comprising meetings from seven U.S. cities, the approach achieves a 10% average F1-score gain and up to 40% improvement in critical issue detection over state-of-the-art methods. Key contributions include: (1) a structure–semantics co-modeling paradigm that jointly leverages hierarchical agenda layouts and domain semantics; and (2) a lightweight, interpretable probabilistic inference architecture specifically designed for governmental text analytics.

Technology Category

• Natural Language Processing: Discourse, Pragmatics & Argument Mining
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Local governments around the world are making consequential decisions on behalf of their constituents, and these constituents are responding with requests, advice, and assessments of their officials at public meetings. So many small meetings cannot be covered by traditional newsrooms at scale. We propose PUBLICSPEAK, a probabilistic framework which can utilize meeting structure, domain knowledge, and linguistic information to discover public remarks in local government meetings. We then use our approach to inspect the issues raised by constituents in 7 cities across the United States. We evaluate our approach on a novel dataset of local government meetings and find that PUBLICSPEAK improves over state-of-the-art by 10% on average, and by up to 40%.

Problem

• Analyzing public remarks in local government meetings
• Utilizing probabilistic framework for meeting structure and linguistic data
• Improving accuracy in detecting constituent issues by 10-40%

Innovation

• Probabilistic framework for public remark discovery
• Utilizes meeting structure and linguistic information
• Improves state-of-the-art performance by 10-40%