Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
From Geographic Location
to Network Location:
The Potential of Big Social Data
Prof. Axel Bruns
ARC Future Fellow
Digital ...
BIG DATA
BIG SOCIAL DATA
(http://siliconangle.com/blog/2011/08/09/twitter-unravel-the-mysteries-of-big-data/)
BIG SOCIAL GEODATA?
Twitter Decahose English-language georeferenced tweets 23 October 2012 to 30 November
2012. (Leetaru e...
BIG SOCIAL GEODATA?
Network map showing locations of users retweeting other users (geocoded Twitter Decahose tweets 23
Oct...
BIG SOCIAL GEODATA?
http://users.humboldt.edu/mstephens/hate/hate_map.html
GEOTAGGING IS UNCOMMON
GEOLOCATION? NETWORK LOCATION!
• Account information available from Twitter API:
– Description (free text)
– Location (fre...
MAPPING A NATIONAL TWITTERSPHERE
• Account information selected:
– Description: mentions of Australian terms, top location...
Education
Agriculture
Literature
Adelaide / SA
Food
Wine
Beer
Parenting
Mums PR
Netizens
Marketing
Investing
Real Estate
H...
ACCOUNTS CREATED IN 2006
ACCOUNTS CREATED IN 2007
ACCOUNTS CREATED IN 2008
ACCOUNTS CREATED IN 2009
ACCOUNTS CREATED IN 2010
ACCOUNTS CREATED IN 2011
ACCOUNTS CREATED IN 2012
ACCOUNTS CREATED IN 2013
AFL GRAND FINAL
Q&A
BACK TO GEOLOCATION?
NETWORK LOCATION  GEOLOCATION?
• Inferring (typical) geolocation:
– More local following than intercity/state/country fol...
LIMITATIONS
• Twitter API policies:
– Pursuit of short-term goals, not long-term strategies
– Ill-conceived push to raise ...
http://mappingonlinepublics.net/
@snurb_dot_info
@jeanburgess
@tsadkowsky
@petamitchell
@flxvctr
@socialmediaQUT – http://...
Upcoming SlideShare
Loading in …5
×

From Geographic Location to Network Location: The Potential of Big Social Data

1,013 views

Published on

Invited presentation at the Pivotal 2015 International Executive Summit, Brisbane, 30 June 2015.

Published in: Social Media
  • Be the first to comment

From Geographic Location to Network Location: The Potential of Big Social Data

  1. 1. From Geographic Location to Network Location: The Potential of Big Social Data Prof. Axel Bruns ARC Future Fellow Digital Media Research Centre Queensland University of Technology Brisbane, Australia a.bruns@qut.edu.au – @snurb_dot_info – http://mappingonlinepublics.net/
  2. 2. BIG DATA
  3. 3. BIG SOCIAL DATA (http://siliconangle.com/blog/2011/08/09/twitter-unravel-the-mysteries-of-big-data/)
  4. 4. BIG SOCIAL GEODATA? Twitter Decahose English-language georeferenced tweets 23 October 2012 to 30 November 2012. (Leetaru et al., 2013 – http://firstmonday.org/article/view/4366/3654)
  5. 5. BIG SOCIAL GEODATA? Network map showing locations of users retweeting other users (geocoded Twitter Decahose tweets 23 October 2012 to 30 November 2012) .(Leetaru et al., 2013 – http://firstmonday.org/article/view/4366/3654)
  6. 6. BIG SOCIAL GEODATA? http://users.humboldt.edu/mstephens/hate/hate_map.html
  7. 7. GEOTAGGING IS UNCOMMON
  8. 8. GEOLOCATION? NETWORK LOCATION! • Account information available from Twitter API: – Description (free text) – Location (free text) – Follower network – Twitter join date – Interface language – Interface timezone – Key stats (# followers, followees, tweets, etc.) • Limitation: – Not available for ‘protected’ accounts (~3.5%)
  9. 9. MAPPING A NATIONAL TWITTERSPHERE • Account information selected: – Description: mentions of Australian terms, top locations – Location: mentions of Australia, top locations – Interface timezone: one of eight Australian state timezones
  10. 10. Education Agriculture Literature Adelaide / SA Food Wine Beer Parenting Mums PR Netizens Marketing Investing Real Estate Home Business Sole Traders Self-Help HR / Support Followback Urban Media Utilities Advertising Business Fashion Beauty Arts Cinema Journalists Politics Hard RightLeftists News CyclingTalkback Music TV V8s UFC NRL AFL Football Horse Racing Cricket NRU Celebrities Hillsong Perth Pop Media Teen Idols Cody Simpson THE AUSTRALIAN TWITTERSPHERE ~140k Australian accounts with degree > 1000, as of Sep. 2013 (of a total 2.8m accounts found)
  11. 11. ACCOUNTS CREATED IN 2006
  12. 12. ACCOUNTS CREATED IN 2007
  13. 13. ACCOUNTS CREATED IN 2008
  14. 14. ACCOUNTS CREATED IN 2009
  15. 15. ACCOUNTS CREATED IN 2010
  16. 16. ACCOUNTS CREATED IN 2011
  17. 17. ACCOUNTS CREATED IN 2012
  18. 18. ACCOUNTS CREATED IN 2013
  19. 19. AFL GRAND FINAL
  20. 20. Q&A
  21. 21. BACK TO GEOLOCATION?
  22. 22. NETWORK LOCATION  GEOLOCATION? • Inferring (typical) geolocation: – More local following than intercity/state/country following?  Can we infer your location from that of your followees? – More discussion of local than non-local issues?  Can we infer your location from your typical topics?  Can we infer your location from your network’s topics? • Combining network and geographic location data: – How does information travel across the network? – Does geographic location affect information flows here?
  23. 23. LIMITATIONS • Twitter API policies: – Pursuit of short-term goals, not long-term strategies – Ill-conceived push to raise revenue through data sales – Counterproductive relationship with research community – Data access shaped to privilege certain limited methods  Most ‘big data’ Twitter research conducted by Twitter, Inc. and commercial research institutes
  24. 24. http://mappingonlinepublics.net/ @snurb_dot_info @jeanburgess @tsadkowsky @petamitchell @flxvctr @socialmediaQUT – http://socialmedia.qut.edu.au/ @qutdmrc – https://www.qut.edu.au/research/dmrc This research is funded by the Australian Research Council through Future Fellowship and LIEF grants FT130100703 and LE140100148.

×