Social Construction of Urban Space: Using LLMs to Identify Neighborhood Boundaries From Craigslist Ads

TL;DR

This study uses large language models and geospatial analysis to explore how Craigslist rental ads reflect social constructions of neighborhood boundaries in Chicago, revealing conflicts, claims, and reputation management in urban space.

Contribution

It introduces a novel approach combining NLP and geospatial analysis to understand social boundary construction through unstructured rental listing data.

Findings

Identifies three patterns of neighborhood boundary claims

Reveals contested and negotiated urban space definitions

Shows correlation between listing language and spatial location

Abstract

Rental listings offer a window into how urban space is socially constructed through language. We analyze Chicago Craigslist rental advertisements from 2018 to 2024 to examine how listing agents characterize neighborhoods, identifying mismatches between institutional boundaries and neighborhood claims. Through manual and large language model annotation, we classify unstructured listings from Craigslist according to their neighborhood. Further geospatial analysis reveals three distinct patterns: properties with conflicting neighborhood designations due to competing spatial definitions, border properties with valid claims to adjacent neighborhoods, and "reputation laundering" where listings claim association with distant, desirable neighborhoods. Through topic modeling, we identify patterns that correlate with spatial positioning: listings further from neighborhood centers emphasize different amenities than centrally-located units. Natural language processing techniques reveal how definitions of urban spaces are contested in ways that traditional methods overlook.