The Influence of the Built Environment on Small Town Image:A Case Study of Tianjin

XU Weixing; LIU Beibei; ZENG Peng; SHI Linfeng; SUN Zongyao

doi:10.3969/j.issn.1000-0232.2025.01.007

Abstract：Small towns are an important link between urban and rural areas.The effective construction of small towns in line with local population needs is a catalyst for urbanization and rural revitalization.A case study based on characteristic towns in Tianjin was carried out.Public data of small towns,including pictures,texts,and check-ins,were collected by Python.The spatial emotion and image genes of small towns were measured using a deep learning method,with the variables of image uniqueness and image intensity.Based on multi-source heterogeneous data,five new objective and measurable dimensions were built using Lynch's classic five elements of"Path,Edge,District,Node and Landmark".They were used to build an explanatory variable system incorporating positive emotions and socioeconomic attributes,which covered 17 indexes.The influence of the concrete built environment on small-town image was investigated through hierarchical linear regression,a moderating effect model,and a geographically weighted model.Results showed that:(1)In terms of the uniqueness of small towns,cultural customs accounted for a relatively high proportion of responses in nine types of image genes.The uniqueness of Balitai was remarkable.Positive emotions were shown to moderate the relationship between the built environment and uniqueness to some extent,including weakening the positive effect of popular tourist attractions and decreasing the negative effect of the mixing degree of POI.However,it had weak moderating effects for the driving distance to urban areas,building density,proportion of open space,and proportion of water areas.(2)In terms of the image intensity of small towns,the image intensity of Yangliuqing,Guanzhuang,Zhongbei,and Xiaying was relatively high.Spatially,there was a core agglomeration pattern adjacent to the west side of the main urban area.The agglomeration of Xiaying and Guanzhuang in the northern Jinbei area was the second highest.As built environmental elements that influence the image intensity of small towns significantly,the area ratio of squares and the density of tourist attractions demonstrated spatial heterogeneity through spatial autocorrelation analysis and a comparison of model performances.The area ratio of squares had a negative effect on the image intensity of small towns,which was mainly low in all small towns.The degree of influence decreased gradually from south to north.There was a positive correlation between the density of tourist attractions and image intensity,and high-value effects accounted for a high proportion of responses.The degree of influence increased from north to south.Based on the research results as well as current observations,some optimization measures which conform to the humanism principle and are in favor of small-town image were proposed by summarizing universal attributions.Based on public social media data,small-town image was quantified according to image uniqueness and image intensity.A new five-dimensional built environmental system composed of multi-source objective heterogeneous big data was established.A model method was proposed to analyze the internal mechanism governing the relationship between the built environment and image in the context of small towns,helping to address the current gap in the research on the digital image of small towns.The research results provide a theoretical basis,methodological support,and practical references for the improvement of small-town image.