Invisible Walls in Cities: Leveraging Large Language Models to Predict Urban Segregation Experience with Social Media Content

2025-02-17

AI Summary

Urban social segregation manifests through subtle, lived experiences that resist detection by conventional methods; meanwhile, social media reviews—though abundant—pose significant modeling challenges due to scale, ambiguity, and heterogeneous perspectives. Method: We propose Reflective LLM Coder, the first theory-grounded, reflective prompting strategy that automatically generates transferable, multidimensional codebooks (e.g., cultural resonance, accessibility, community engagement). Building upon this, we introduce the RE’EM framework, which synergistically integrates large language model (LLM) reasoning and embedding capabilities for cross-city modeling of segregation experiences. Contribution/Results: Experiments show a 22.79% improvement in R² and a 9.33% reduction in MSE over baselines. The generated codebooks demonstrate cross-city generalizability across three metropolitan areas. A user study confirms that our approach significantly enhances efficiency in assessing point-of-interest (POI) social inclusivity—validating both methodological rigor and practical utility.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Understanding experienced segregation in urban daily life is crucial for addressing societal inequalities and fostering inclusivity. The abundance of user-generated reviews on social media encapsulates nuanced perceptions and feelings associated with different places, offering rich insights into segregation. However, leveraging this data poses significant challenges due to its vast volume, ambiguity, and confluence of diverse perspectives. To tackle these challenges, we propose using Large Language Models (LLMs) to automate online review mining for segregation prediction. We design a Reflective LLM Coder to digest social media content into insights consistent with real-world feedback, and eventually produce a codebook capturing key dimensions that signal segregation experience, such as cultural resonance and appeal, accessibility and convenience, and community engagement and local involvement. Guided by the codebook, LLMs can generate both informative review summaries and ratings for segregation prediction. Moreover, we design a REasoning-and-EMbedding (RE'EM) framework, which combines the reasoning and embedding capabilities of language models to integrate multi-channel features for segregation prediction. Experiments on real-world data demonstrate that our framework greatly improves prediction accuracy, with a 22.79% elevation in R2 and a 9.33% reduction in MSE. The derived codebook is generalizable across three different cities, consistently improving prediction accuracy. Moreover, our user study confirms that the codebook-guided summaries provide cognitive gains for human participants in perceiving POIs' social inclusiveness. Our study marks an important step toward understanding implicit social barriers and inequalities, demonstrating the great potential of promoting social inclusiveness with AI.

Problem

• Predict urban segregation using social media content.
• Automate review mining with Large Language Models.
• Improve segregation prediction accuracy with AI frameworks.

Innovation

• Uses Large Language Models for urban segregation analysis
• Develops Reflective LLM Coder for social media insights
• Implements RE'EM framework for enhanced prediction accuracy