Skip to main content

Research Repository

See what's under the surface

Advanced Search

A data driven approach to mapping urban neighbourhoods

Brindley, Paul; Goulding, James; Wilson, Max L.

Authors

Paul Brindley

James Goulding

Max L. Wilson



Abstract

Neighbourhoods have been described by the UK Secretary of State for Communities and Local Government as the “building blocks of public service society”. Despite this, difficulties in data collection combined with the concept’s subjective nature have left most countries lacking official neighbourhood definitions. This issue has implications not only for policy, but for the field of computational social science as a whole (with many studies being forced to use administrative units as proxies despite the fact that these bear little connection to resident perceptions of social boundaries). In this paper we illustrate that the mass linguistic datasets now available on the internet need only be combined with relatively simple linguistic computational models to produce definitions that are not only probabilistic and dynamic, but do not require a priori knowledge of neighbourhood names.

Publication Date Jan 1, 2014
Peer Reviewed Peer Reviewed
Series Title SIGSPATIAL '14
APA6 Citation Brindley, P., Goulding, J., & Wilson, M. L. (2014). A data driven approach to mapping urban neighbourhoods
Keywords neighbourhoods, spatial data mining, vernacular geography
Publisher URL http://doi.acm.org/10.1145/2666310.2666473
Copyright Statement Copyright information regarding this work can be found at the following address: http://eprints.nottingh.../end_user_agreement.pdf
Additional Information Published in: Proceedings of the 22nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. ACM, 2014, ISBN 9781450331319, p. 651. doi: 10.1145/2666310.2666473

Files

p437-brindley.pdf (383 Kb)
PDF

Copyright Statement
Copyright information regarding this work can be found at the following address: http://eprints.nottingham.ac.uk/end_user_agreement.pdf





You might also like



Downloadable Citations

;