Socioeconomic and functional zoning characterization in a city: a clustering approach

The interpretation of urban landscapes, functional zoning, and socioeconomic stratification are crucial factors influencing our perception of areas' characteristics and dynamics, offering essential insights for sustainable urban planning and design. While traditional survey data are time-consuming and could result in poor samples, data available via the Internet provides a new way to identify socioeconomic conditions and spatial functions in a city. However, existing methods often face limitations: approaches using only remote imagery or POI (Point of Interest) data fail to capture nuanced socioeconomic features, while social sensing data alone lacks morphological context. To address these challenges, this study presents a data-driven framework integrating POI data and census information, analyzed using an unsupervised machine learning algorithm (K-Means), to recognize patterns of different socioeconomic levels and functional zones in Lisbon city. Using census data, four distinct types of socioeconomic zones were identified and classified based on specific demographic and economic characteristics. Additionally, the temporal analysis of POI data revealed varying functionalities across zones of different socioeconomic levels, with distinct activity patterns observed at different times of the day and the week. Zones of higher socioeconomic levels exhibited more recreational activity during weekends, while lower socioeconomic areas predominantly supported business-related services. By providing a comprehensive view of the interconnection between socioeconomic levels and functional dynamics, this study offers actionable insights into Lisbon's urban structure. Furthermore, the scalable framework demonstrates significant potential for application in other urban contexts, with implications for urban planning, resource allocation, and social equity. • Identification of socioeconomic status and functional zoning in a city based on the analysis of spatial data • Analysis of patterns between urban functional areas and their effects on socioeconomic inequalities and vice-versa • The methodology can be applied to various cities provided that the data is available • Relevant information to support urban planning and management