Angela Zoss and I are using data from Larry Carin's research group at Duke University to try to visualize and navigate a network of scholars and publication forums from the Scholars@Duke site. Specifically, Piyush Rai and Changwei Hu supplied us with data on author similarity based on new algorithms developed in Larry Carin's lab.
I've been testing this by running a simple Python server in the developement directory:
python -m SimpleHTTPServer 8888
and then navigating to http://127.0.0.1:8888/scholars_network_prototype.html or http://127.0.0.1:8888/forums_network_prototype.html
NOTE: This is just prototype code. Feel free to take it and play, but it's just unsupported research code for exploring how one might navigate this data.
For the interactive visualization I pulled from a lot of d3.js example code. Notably, Simon Raper's page on a-z of extra features for the d3 force-directed layout and Mike Bostock's 2nd zoom to bounds example. I also used Justin Palmer's d3-tip library for tooltips, and took a lot of style and layout inspiration for the info panels from the D3plus project/product.