A Method for Analyzing Urban Socio-spatial Structure Based on Multi-source Data - Taking Shenyang as An Example

Urban socio-spatial structure reflects the spatial manifestation of urban social differentiation and is one of the key research topics in urban social geography. Traditional methods of measuring urban socio-spatial structure rely on census data collected at the level of the street office and employ factor analysis and cluster analysis techniques. However, these methods face the challenges of a mismatch between administrative boundaries and social space boundaries, and low spatial resolution due to the scale of data aggregation. To overcome these limitations, this paper proposes a novel method to estimate socio-spatial structure using mobile phone signaling data, housing price data, and enterprise location data. The method first identifies the social attributes and mobility patterns of the working and residential populations based on the signaling data, then cross-analyses the location of employment centers derived from the enterprise location data and the income level information inferred from the housing price data, and finally produces a four-dimensional result including residential density, age, income level, and employment type. The application of this method in Shenyang reveals a typical “circle + sector” structure as illustrated in Figure below. The results show that young people with high income levels and office jobs are mainly concentrated in areas near Shenyang Station, Shenyang North Station, and Northeastern University, as well as along the south bank of the Hun River. And older people with low income levels and mixed types of jobs are mainly distributed in the eastern part of Second Ring Road and the southwestern part of the First-Second Ring Road area. This method offers a more precise characterization of the urban socio-spatial structure than conventional approaches and can facilitate more tailored interventions for urban spatial planning.