Evolution of a Revolution: Visualizing Millions of Iran Tweets

Computing with Social Structures

Simply tracking the volume of various phrases gives us a sense of what is happening on the street, literally and figuratively. But that signal is but a shadow of a far more complex and intricate reality, an interwoven web of individuals and actions.

Twitter allows these social structures to become data structures by means of the "RT" convention. And this in turn allows us to perform extremely powerful computations on the social structures that underlie the flow of information.

Network layout algorithms are a familiar, powerful, and fascinating example. They self-organize in your computer to reveal self-organization in the real world. And that is exactly the kind of tool we need to test our hypothesis about #cnnfail.

The plot below shows the network of people who re-tweeted mentions of IP proxies, with those who had tweeted earlier about #cnnfail highlighted. We see not only significant overlap among the people involved but also a considerable structure in the relationships between them. We have captured a real community at the moment of its birth.

Remember this as you look at the next plot below. Here, we see the re-tweet network that formed around the top five Iranian tweets. Its structure shows a very different phenomenon, capturing the emergence not of a community but of an elite. Despite massive interest, or perhaps because of it, most people did not discover more than one of the top Iranians. The network simply grew faster than the information could naturally propagate. But a small inner circle did succeed in identifying core sources of information.

The final plot below shows yet another community structure, as well as a new algorithmic technique. This plot does not show the emergence of a new community but rather shows the appropriation of a new topic by mature political factions. This re-tweet network has formed around Iranian tweets that mention Obama. Using graph theory, we can computationally extract the sub-communities and then use that information to color the network. The large blue mass on the right is the conservative Twittersphere, while the other structures are a less-organized collection of mainstream or progressive news outlets.