Cover art for AI-Supported Objection Management in Public Participation: Concept, Prototype and Evaluation in the Context of Infrastructure Projects
Technology

AI-Supported Objection Management in Public Participation: Concept, Prototype and Evaluation in the Context of Infrastructure Projects

Written by Jonathan Matthei, Johannes Maas, Maurice Wischum, Sven Mackenbach, Katharina Klemt-Albert

54:08Eric26 tracks

Narration by Research @research
Listen free

About

This study explores how large language models can enhance the efficiency of processing public objections in complex infrastructure planning projects. It evaluates a prototype designed to assist administrative workflows while maintaining human-in-the-loop oversight. From mdpi.com/2571-5577/9/6/107, licensed under CC BY 4.0. Audio narration of the original text.

#artificial intelligence#infrastructure#public participation#large language models#governance#administrative efficiency
Read the full text →

Contents

  1. 1Abstract1:50
  2. 21. Introduction0:01
  3. 31.1. Initial Situation3:16
  4. 41.2. Research Aim and Procedure1:35
  5. 52. Theoretical Foundation0:01
  6. 62.1. Definition, Background, and Classification of Artificial Intelligence3:16
  7. 72.2. Background of Large Language Models and Retrieval-Augmented Generation3:09
  8. 82.3. Background on Vector Databases and the Embedding Process2:57
  9. 92.4. Background to Prompt Engineering1:25
  10. 103. Use Cases for an AI-Supported Objection Management System3:46
  11. 113.1. Thematic Pre-Sorting0:47
  12. 123.2. Text Summary0:38
  13. 133.3. Text Similarity Check0:40
  14. 143.4. Response Text Generation0:46
  15. 154. AI-Supported Objection Management System0:54
  16. 164.1. AI-Supported Distribution and Categorization of Objections0:04
  17. 174.1.1. Architecture2:44
  18. 184.1.2. Prototypical Implementation0:55
  19. 194.2. Generation of Suggested Responses0:02
  20. 204.2.1. Architecture1:44
  21. 214.2.2. Prototypical Implementation1:08
  22. 224.3. User Interface1:19
  23. 235. Validation0:01
  24. 245.1. Methodology4:37
  25. 255.2. Validation Results4:02
  26. 266. Summary, Discussion, and Outlook12:18