Our new paper on understanding and modeling cities through texts in the age of AI is published in Nature Cities. It discusses how texts can be leveraged to understand cities and people for applications from urban planning to disaster management.
Urban researchers now have access to vast amounts of textual data—from social media and news to planning documents and property listings. These textual data provide important information about the activities of people and organizations in urban environments. Meanwhile, recent advancements in computational tools, including large language models, have expanded our ability to analyze textual data. This article explores how these tools are reshaping the ways we analyze, understand, and theorize the city through text. By outlining key developments, applications, and challenges, it argues that text is no longer a ‘fringe resource’ but a central component in urban analytics with the potential to connect quantitative and qualitative researchers.
Full paper: https://www.nature.com/articles/s44284-025-00314-x [pdf]