European Communication Conference (ECREA) 2014
Mapping a National Twittersphere: A ‘Big Data’ Analysis of Australian Twitter User Networks
Axel Bruns, Darryl Woodford, Troy Sadkowsky, and Tim Highfield
Twitter research to date has focussed mainly on the study of isolated events, as described for example by specific hashtags or keywords relating variously to elections (Larsson & Moe, 2012), natural disasters (Mendoza et al., 2010), entertainment (Highfield et al., 2013) and sporting events (Bruns et al., 2014), and other moments of heightened activity in the network. This limited focus is determined in part by the limitations placed on large-scale access to Twitter data by Twitter, Inc. itself. By contrast, only a handful of studies – usually by researchers associated with commercially funded research organisations or with Twitter, Inc. itself – have utilised the Twitter ‘firehose’ or similar more comprehensive sources of data to explore broader patterns of traffic flows or follower connections on the platform (e.g. Leetaru et al., 2013).
This project builds on a long-term, large-scale analysis of the global Twitter userbase which has managed to identify within the over 725 million global registered Twitter accounts some 2.5 million Australian accounts (by matching profile details such as location, description, and timezone against a set of relevant criteria). Further, we analysed the follower/followee connections of these 2.5 million accounts and from this developed a first comprehensive map of account relationships within the Australian Twittersphere. In-depth network analysis of this map reveals the existence of a range of clusters of especially tightly interconnected users, linked to each other by other accounts acting as bridges between the clusters. In turn, qualitative exploration of the leading account’s profiles in each cluster provides an indication of the various areas of thematic focus which have determined the formation of these clusters, and their association with other clusters in the same network vicinity. Further correlation with other relevant profile data (including the creation date for each account, its level of tweeting activity, and the date of the account’s last tweet) offers additional opportunities to trace the emergence and growing complexity of the Australian Twittersphere over time, from the earliest adopters of the platform to its most recent users, and to filter the overall network for the most active and most persistent users.
This study represents the first ever comprehensive investigation of the development of a national Twittersphere as an entity in its own right. While the global nature of Twitter as a social media platform means that Australian accounts will also be connected with their counterparts in other countries, it is still to be expected that shared interests and identity lead to the majority of connections between accounts to occur within the same national user population, and our analysis of these connection patterns provides an important indicator of the themes around which these connections crystallise, as well as of the longitudinal development of these clusters of interests.
Bruns, A., Weller, K., & Harrington, S. (2014). Twitter and Sports: Football Fandom in Emerging and Established Markets. In K. Weller, A. Bruns, J. Burgess, M. Mahrt, & C. Puschmann (Eds.), Twitter and Society (pp. 263–280). New York: Peter Lang.
Highfield, T., Harrington, S., & Bruns, A. (2013). Twitter as a Technology for Audiencing and Fandom: The #Eurovision Phenomenon. Information, Communication & Society, 16(3), 315–39. doi:10.1080/1369118X.2012.756053
Larsson, A.O., & Moe, H. (2011). Studying Political Microblogging: Twitter Users in the 2010 Swedish Election Campaign. New Media & Society, 14(5). doi:10.1177/1461444811422894
Leetaru, K., Wang, S., Cao, G., Padmanabhan, A., & Shook, E. (2013). Mapping the Global Twitter Heartbeat: The Geography of Twitter. First Monday, 18(5). doi:10.5210/fm.v18i5.4366
Mendoza, M., Poblete, B., & Castillo, C. (2010) Twitter under Crisis: Can We Trust What We RT? Paper presented at Social Media Analytics, KDD '10 Workshops, Washington, DC, 25 July 2010. Available from: http://research.yahoo.com/files/mendoza_poblete_castillo_2010_twitter_terremoto.pdf