University of North Texas
The Democratic Right to Comprehend Public Data: Enabling Non-technical Users to Discover, Question, and Analyze Open Data Using Language Model Agents
Abstract
dc:descriptionIn recent years, a growing emphasis on data transparency has led our public institutions to publish increasingly extensive datasets. However, the progression from historically simpler social structures to today's complex, information-dense environments has dramatically increased both the volume and complexity of public data. Consequently, individuals, experts, and organizations face substantial challenges and costs in digesting and understanding this information. This complexity effectively creates technical barriers for a broad range of stakeholders in society who are not data and programming experts. At the individual level, it compromises their ability to practice informed citizenship, and at the organizational level, it slows the pace of legislation and administrative measures, thus creating the paradox of modern public data: information that is open and available, yet remains functionally inaccessible to those without specialized technical skills. Our research utilizes advances in artificial intelligence to address this paradox. The proposed solution is an open-source framework offering an intuitive interface where non-technical users can ask questions about public data in plain language. Behind this interface, multiple large language model agents work to access open dataset repositories such as data.gov to find, retrieve, and analyze the relevant data. This framework is extensible to any online or offline dataset repository in order to remove programming and data analysis barriers for a wide range of individuals and organizations. The framework's data-driven response is multi-modal and contains the explanation of the data analysis experiments as well as the supporting data. This work aims to democratize the understanding of public data, empower data-informed civic participation, and improve administrative agility in public and private organizations through data-driven decision making. It is a step toward a future where the mitigated complexity of public data is recognized as a fundamental democratic right and an essential factor for meaningful participation in modern society.
Degree
thesis:*- Grantor dc:publisher
- University of North Texas
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Montazeri, Sina
- Contributors dc:contributor
-
- Jesus Quevedo-Torrero
- Haihua Chen
- Yunhe Feng
Subjects
dc:subject × 5Rights
dc:rights- Statement dc:rights
-
- Public
- Montazeri, Sina
- Copyright
- Copyright is held by the author, unless otherwise noted. All rights Reserved.
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
-
local-cont-no: submission_4374
https://digital.library.unt.edu/ark:/67531/metadc2443387/
ark: ark:/67531/metadc2443387