Collective sensing of evolving urban structures: From activity-based to content-aware social monitoring

Eszter Bokányi, Zsófia Kallus and István Gódor
Additional contact information
Zsófia Kallus: Eötvös Loránd University, Hungary; Ericsson Research, Hungary

Environment and Planning B, 2021, vol. 48, issue 1, 115-131

Abstract: Cities are constantly evolving complex systems. Detection methods of land use distribution have to keep pace with their rapidly changing landscapes. While traditional analysis relies on surveys and census data typically refreshed at most yearly, the widespread use of mobile devices allows cell phone activity measurements to be used as sensors for the functional clustering of urban districts. These activity-based proprietary measurements are recently complemented by publicly available geosocial network records that even enable content-aware analysis. As a bridge between separate methods, in this work we analyze the relation of conversation content and functional spatial clusters of cities using a double dataset approach. We look at the differentiating power of the content of local conversations in activity-driven land use detection based on mobile phone records. In addition to intra-city analysis of three metropolises, we present a comparative study of London, New York City, and Los Angeles sharing the common language of English, but having very different cultural backgrounds. We show that the share of words with a similar temporal pattern to that of local mobile activities is different across cities, as well as between functional clusters. We find that the conversational content can effectively differentiate both functional clusters of a single city, and similar locations of the same function across many cities, like business areas that otherwise have a common temporal heartbeat. Moreover, we identify words related to activity types as the most important features emerging from the content-based, data-driven classification.

Keywords: Urban clustering; urban land use detection; text analysis; machine learning; geospatial data; geosocial data (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.sagepub.com/doi/10.1177/2399808319848760 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sae:envirb:v:48:y:2021:i:1:p:115-131

DOI: 10.1177/2399808319848760

Access Statistics for this article

More articles in Environment and Planning B
Bibliographic data for series maintained by SAGE Publications ().