Public Perception of City Image Hotspots Based on Social Media: A Case Study of Nanjing, China

Abstract

City image is an important element in the design of urban features. Big data from social media has become a new way to perceive city image characteristics. Taking Nanjing (China) as a case study, we use social media data from the “Little Red Book” and “Sina Weibo” (similar to Twitter in China) and employ deep learning methods, multisource data semantic analysis, sDNA, and geographically weighted regression to analyze three aspects: the spatial distribution of hotspots, perception characteristics, and influencing factors. The hotspots are located in Nanjing’s old city, centered on Confucius Temple, and extend outward in a circular pattern. Regarding perception characteristics, the key characteristics that define the old city are “heritage monuments”, “long history”, and “culture”. The high functional perception area converged around the central urban area. The spatial distributions of closeness and betweenness under traffic perception showed opposite trends; emotional perception was mainly positive. Compared with visual and emotional perception, functional and traffic perception have greater impacts on city image hotspots. Our study constructs a model of city image perception from a new perspective. This approach bridges the gap that traditional city image research focuses only on objective environment descriptions and lacks subject-object relationship analysis, which can provide scientific value for decision makers in urban design management.