Geo-temporal Twitter demographics

ABSTRACT This paper seeks and uses highly disaggregate social media sources to characterize Greater London in terms of flows of people with modelled individual characteristics, as well as conventional measures of land use morphology and night-time residence. We conduct three analyses. First, we use the Shannon Entropy measure to characterize the geography of information creation across the city. Second, we create a geo-temporal demographic classification of Twitter users in London. Third, we begin to use Twitter data to characterize the links between different locations across the city. We see all three elements as data rich, highly disaggregate geo-temporal analysis of urban form and function, albeit one that pertains to no clearly defined population. Our conclusions reflect upon this severe shortcoming in analysis using social media data, and its implications for progressing our understanding of socio-spatial distributions within cities.

[1]  Paul A. Longley,et al.  Towards Real‐Time Geodemographics: Clustering Algorithm Performance for Large Multidimensional Spatial Databases , 2010, Trans. GIS.

[2]  Paul A. Longley,et al.  The Geotemporal Demographics of Twitter Usage , 2015 .

[3]  Ana-Maria Popescu,et al.  A Machine Learning Approach to Twitter User Classification , 2011, ICWSM.

[4]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[5]  Matthew Zook,et al.  Social Media and the City: Rethinking Urban Socio-Spatial Inequality Using User-Generated Geographic Information , 2015 .

[6]  Matthew Zook,et al.  Offline Brews and Online Views: Exploring the Geography of Beer Tweets , 2014 .

[7]  Richard Webber,et al.  Geodemographics, GIS and Neighbourhood Targeting , 2005 .

[8]  Michael Batty,et al.  Space, scale, and scaling in entropy-maximising , 2010 .

[9]  Norman M. Sadeh,et al.  The Livehoods Project: Utilizing Social Media to Understand the Dynamics of a City , 2012, ICWSM.

[10]  Ate Poorthuis,et al.  Follow thy neighbor: Connecting the social and the spatial networks on Twitter , 2015, Comput. Environ. Urban Syst..

[11]  Wendy Liu,et al.  Homophily and Latent Attribute Inference: Inferring Latent Attributes of Twitter Users from Neighbors , 2012, ICWSM.

[12]  Matthew Zook,et al.  Social Media and the City: Rethinking Urban Socio-Spatial Inequality Using User-Generated Geographic Information , 2015 .

[13]  Daniele Quercia,et al.  Talk of the City: Our Tweets, Our Community Happiness , 2012, ICWSM.

[14]  M. Batty.  The New Science of Cities , 2013 .

[15]  P. Rees,et al.  Creating the UK National Statistics 2001 output area classification , 2007 .

[16]  Paul Knox,et al.  Behavioral Problems in Geography Revisited , 1983 .

[17]  P. Longley,et al.  Ethnicity and Population Structure in Personal Naming Networks , 2011, PloS one.

[18]  Michael Batty,et al.  Fractal Cities: A Geometry of Form and Function , 1996 .

[19]  Rizal Setya Perdana.  What is Twitter , 2013 .

[20]  J. H. Ward.  Hierarchical Grouping to Optimize an Objective Function , 1963 .

[21]  M. Goodchild.  Citizens as sensors: the world of volunteered geography , 2007 .

[22]  Nick Malleson,et al.  The Classification of Space-Time Behaviour Patterns in a British City from Crowd-Sourced Data , 2013, ICCSA.