Geography Reference
In-Depth Information
actively and frequently engaged in interactions. De Longueville et al. (
2009
)
analysed the Twitter social network during moments of environmental crises.
They identified a segmentation process of the user group into distinct types of
communicators that handle and forward crisis information in different ways.
Several studies are focusing on understanding the demographic characteristics of
the Twitter users at global level (Kulshrestha et al.
2012
), at country level (Mislove
et al.
2011
) or at city level (Adnan and Longley
2013
). Kulshrestha et al. (
2012
)
evaluated the distribution of the Twitter users across the world and the Twitter
adoption rate compared to the socio-economic status of the country
'
s population
such as the Human Development Index (HDI). The results of this study showed a
highly unequal distribution of the Twitter users across countries and a high corre-
lation between the HDI and Twitter adoption rate. Mislove et al. (
2011
) compared
the number of Twitter users with the number of population in US counties and
assessed the ethnicity and gender of the Twitter users. This study revealed that there
exists a significantly larger number of male Twitter users in the US. Furthermore,
sparsely populated areas are underrepresented in terms of tweeting activity.
Adnan and Longley (
2013
) performed a descriptive analysis of Twitter users in
London, Paris and New York. In particular, they studied demographic aspects of
Twitter users such as name, ethnicity and gender to get deeper insights into the
ethnic diversity of the cities
population. The authors assessed also the hourly
twitter activity and identified two tweeting activity peaks in London: between
10 a.m. and 11 a.m. and between 7 p.m. and 11 p.m. The geographic distribution
of the day and night tweeting activity was not assessed. In a related article, Adnan
et al. (
2013
) investigated the spatial distribution of tweets from different ethnic
groups across London. They separated the messages into day-time and night-time
tweets in order to gain insights in activity patterns across London.
This work aims at describing the socio-demographic characteristics of residents
in locations with high tweet counts. The question is whether socio-demographic
variables predict numbers of Twitter messages and whether the distribution of
tweets provides insights into the demographic group of Twitter users. The specific
contribution of this work is the exploration of representations and analyses for
investigating who Twitter users in London are.
'
Data and Methods
The Twitter community produces tremendous amounts of data on a daily basis. To
access these data for development and research purposes, Twitter offers a Stream-
ing Application Programmable Interface (API).
1
Via this API, a subset of
georeferenced tweets can be accessed at any point in time. To access the API in a
1
Twitter Streaming API—
https://dev.twitter.com/docs/streaming-apis
(2013-12-05).
