In July 2008 CrunchBase released an API for accessing their dataset on web companies. We used this data to draw graphs in order to visualize connections among them.

Available graphs

pdf   png Companies

In this graph companies are considered connected if some of the same key people have worked in both. Tip: use find in your PDF reader to quickly locate a company.

pdf   png People

Lines appear between people when they are associated with the same companies. Marc Andreessen and Gina Bianchini are co-founders of Ning.

png Glimpse of raw data

Before calculating maximum spanning tree and unipartite projection. This is the data we started with.


Details

The graphs were made by HUT research student Mikko Kivelä and University of Tampere grad student Bemmu Sepponen. A simple PHP script was used to create Himmeli vertex and edge data (vertices.txt, edges.txt) from CrunchBase API output. This resulted in the messy graph, which was used as a starting point to extract company and people connections with unipartite projection and then simplifying these resulting graphs by creating a maximum spanning tree.

The graph image files are quite large. If you post them on a high-traffic site, please make local copies so that our server doesn't die. Feel free to use them as you like!

Bonus: Google Number

See how many people are needed in between to connect a company to Google: google_number.txt