Visualising the #Data15 Community
This year was my second visit to the London Tableau Conference and it was a cracking event once again. Loads of new ideas, clever presenters, some great demos and a few beers makes for an inspiring couple of days.
The hashtag #Data15 was busy throughout the conference and I noticed that Jonathan MacDonald of The Information Lab had started a rapidly growing Twitter list of everybody who’d used it. I can’t resist a Twitter network visualisation, so dusted off Python and set to work…
Starting from Jonathan’s Twitter list, I pulled a dataset of every follower and following relationship for its 900+ members. Even capping accounts with very large numbers of followers at 10,000, this is still 1.7 million individual relationships. Network analysis gets big quickly!
I also pulled some details for each account, like the user’s location, description and total follower count.
Tableau Public to build a simple user browser (I couldn’t not use some Tableau for this project)
And the result…
Here’s the #data15 community. If you’re on the original Twitter list, or you’re followed by at least ten people on that list, then you’re in here somewhere.
Twitter users who follow each other, cluster more closely together in the vis and using Gephi we can automatically identify and colour code a few distinct communities who used the #data15 hashtag. Data15 extends much wider than just the usual Tableau social media suspects, who are all clustered together in the top-left corner.
Broadly, the communities look to me like Tableau and Data Visualisation, Big Data, Data Science, News Sources and Not for Profit. There are also a few journalists scattered across the vis, identified in dark blue.
Who should you follow? This Tableau dashboard lets you browse the network and find yourself some Data Vis, or Big Data inspiration.
If you liked these visualisations, or have other ideas for Twitter social network datasets, then do drop me a line.