Sociodigital discourse in the design and planning of rapidly urbanizing industrial cities

This study examines the public sentiment toward Suzhou urban amenities through an analysis of bottom-up discourses on the Chinese social media platform Zhihu. A dataset of over six thousand comments was processed using a pre-trained Chinese BERT-based sentiment classification model to identify emotional tone, recurring urban concerns, and thematic clusters including transport, housing, public space, retail access, and social belonging in the context of the city of Suzhou. Findings reveal that mobility challenges, uneven access to everyday amenities, and tensions around cultural identity generate the highest levels of negative sentiment and social misalignment. These patterns point to a gap in planning protocols, where centralized and efficiency-oriented models, accelerated by rapid industrialization, tend to overlook the lived realities of diverse residents. Discourses on the Zhihu platform offer thick data that complement thin institutional datasets by capturing nuanced, context-rich perspectives at scale and speed. By mapping online expressions of frustration onto real-time diagnostic tools alongside well-established public engagement processes in urban planning, this method can help planners and policymakers to operationalize sociodigital data into generalizable problem-identification and planning methods, leading to a more equitable and adaptive planning process in rapidly urbanizing Chinese industrial cities.