Several years ago, I created a map of scientific collaborations. The attention this map obtained surpassed my wildest expectations; it got published in the scientific and popular press all around the world! I had mainly forgotten about it until I received an email that rekindled my interest in this visualization and I thought it was high time to revisit this visualization.
Unfortunately, scientific papers (and associated data) are closely guarded and only a handful of firms have full access to them. I now work in a very different field, so I lost access to this dataset. But while perusing my Twitter feed, I came across the very active feed of Scimago Lab. Their social media presence and their incredible interactive visualizations convinced me that they might be interested in collaborating. I sent off an email to their founder, Félix de Moya and, lo and behold, he was interested in collaborating. Cool!
Read on for more maps and an overview of the methodology >>
After a bit of back and forth, I spent a weekend programming a tool to draw large geographical graphs. The tool I used a couple of years ago was riddled by projection bugs and was terribly inefficient (and the source was lost when I reformatted an old hard disk), so a rewrite was in order. The idea behind the tool is really simple. It loads up a graph in memory and then goes through the graph edge by edge to draw every edge following the curvature of the earth and projecting each line using the Plate Carré (overlappable on Google Maps) or Eckert III (which I find beautiful).
I shared my tool with Félix and after a trying a good deal of different settings, we were able to compute the following map using the Scopus database (papers published between 2008 and 2012). This map shows the collaboration networks between researchers in different cities. Apart from its aesthetic qualities, the map is useful to illustrate some interesting collaboration patterns. One of the most striking one is the importance of Paris in French science. It seems that every researcher in France collaborates with at least a researcher in Paris. Unsurprisingly, the map also shows quite clearly that the location scientific institutions follows the population density. Also, links between countries and their old colonies are also very obvious.
These following maps were rendered with different color schemes. Click on any map to enlarge.
Because the basic input for this tool is a large collaboration graph between cities, we can also use network analytical tools and methods. One of these methods, the Louvain method (available as a standalone tool or in Gephi) can identify communities in social network graphs. In our case, it can identify different collaboration patterns between cities; these patterns seem to follow linguistic or old colonial lines:
My main preoccupation with this work was mostly aesthetic, but a lot of interesting analytical work could be done. For example, one could illustrate different fields (astronomy, biology, etc.), illustrate different institutions or even animate collaboration patterns during a specific time period, etc. If you think that this could be interesting or useful, please contact Scimago Lab. They’ll be able to answer your questions or even produce custom maps.