打印

作者

Fernanda de Oliveira Capela;Jose Emmanuel Ramirez-Marquez

来源

CITIES,Vol.93,Issue1,Pages 72-83

语言

英文

关键字

City identity;Urban perception;Topic detection;Text analytics;LDA modeling

作者单位

Stevens Institute of Technology, School of Systems and Enterprises, Hoboken, NJ, United States of America;Stevens Institute of Technology, School of Systems and Enterprises, Hoboken, NJ, United States of America

摘要

Understanding subjective aspects of urban life can be a powerful tool for policymakers to improve public well-being by focusing on the peculiarities their metropolis. Cities are as diverse as the humans living in it. Just as any other complex system, a city changes and evolves with time, accumulating experiences and features that are particular to their history. Unmatched memories are what makes people feel connected to a place. This uniqueness of characteristics is what we call the city identity. This paper proposes a method to detect and compare identities of cities using news articles. Revealing the most unique topics for each city can uncover their utmost interests and points of view. We exemplified its use with a set of thirty-six big cities from the United States of America creating a footprint of components containing the city's identity profiles. This method allows the comparison of cities at scale without compromising the quality of the results.