Visualizing Persian public opinion: The Iranian blogosphere map
Really interesting visualization of the social/political landscape in Iran at the moment using computational social network mapping. It’s current enough to reflect public opinion on the day of the election in Iran. Here’s how the map is drawn, according to the creators:
“Our map of the Iranian blogosphere is a social network diagram that shows the relationships among the many kinds of bloggers active in Iran. It combines two analytic techniques to build a visualization that makes this very complex network more easily understood. The first technique is to draw the position of the blogs, each represented by a dot, using what is called a “spring embedder” or “physics model” algorithm. This map uses a particular algorithm known as Fruchterman-Rheingold, after its inventors. Imagine there is a wind trying to blow all the blogs to the edge of the circle, but any two blogs that link to each other are pulled together as though by an invisible spring. This causes groups of densly interconnected blogs to cluster together into “network neighborhoods.” The arrangement of dots on the map, in which the size of each dot reflects the number of other blogs linking to it, shows these network neighborhoods, or “proximity clusters.” The second technique involves statistically grouping blogs according to what they link to over a long period of time, i.e. what websites the bloggers prefer. These groups, which we call “attentive clusters,” are indicated by the colors of the dots. Blogs of the same color tend to link to similar websites.”