城市治理问题的空间识别与影响因素研究——基于武汉城市留言板数据的探索

电子政务平台用户数据为认知城市问题、开展空间治理提供了新路径。研究以武汉市为例，以日常治理问题为对象，基于留言板数据及自然语言处理技术，提出城市治理问题空间识别方法，利用空间自相关和负二项回归分析，探究治理问题空间特征及影响因素。研究发现：(1)治理问题热点区集中于开发区及新旧城区交界，冷点区集中于老城核心地带；(2)社会交互因素上，老年群体在线上治理中存在“失语”现象，高学历群体更重视多元包容类问题表达；(3)地理因素上，人口密度增加并不必然导致城市治理问题增加，为高密度城市实现高质量治理提供依据。

RESEARCH ON SPATIAL IDENTIFICATION AND INFLUENCING FACTORS OF URBAN GOVERNANCE PROBLEMS: AN EXPLORATION BASED ON CITY MESSAGE BOARD DATA OF WUHAN

City message boards and similar e-government platforms are playing an increasingly important role in urban governance. Taking urban everyday governance as the research object, this study classifies urban problems into four types, namely urban safety, diversity tolerance, living comfort, and ecological livability and proposes a method for spatial identification of urban governance problems based on Natural language processing (NLP) technology and Wuhan's city message board data. The findings show that: 1) Four types of governance problems including urban safety, diversity tolerance, living comfort, and ecological livability all have hot spots. 2) Cold spots of comfortable life problems and diversity tolerance problems are witnessed, mainly located in the core area of the old city. 3) In terms of social-interactive factors, there is a significant difference in the discursive power between children and the elderly. 4) In terms of geographic factors, the rise in population density does not necessarily lead to increasing urban governance problems, which provides evidence for the high-quality governance of high-density cities. 5) Owing to the trend of concurrently developing urban construction and urban quality, the effects of environmental factors on the spatial pattern of urban problems are more complicated than a simple linear relationship.