Citation graphs of interconnected communities

As a follow-up to my previous post about the citation graph of the Graph Drawing conference I computed a few more drawings with gephi. This time I have concentrated on visualizing connections between conferences with a similar scope. I improved my "springerlink-crawler". In the current version I first scan all the titles and then I scan for references where the matching is done by title. By this I will also detect references to journal versions if they have the same title. Again the citation graphs are very sparse, since they only reflect citations within the conferences. The data set is very likely not 100% accurate, but my impression is that it is good enough to illustrate a few interesting things.

GD and WG

Graph drawing is a smaller part of the Workshop on Graph-Theoretic Concepts in Computer Science (WG). The graph drawing conference (GD) also targets geometric graph theory. As a big difference, GD is more self-referential, whereas there are considerably less links within the WG citation network. The picture shows the combined citation network of both conferences. Papers appeared in WG are colored yellow, GD papers are drawn red. Many of the WG papers are isolated nodes or show up in very small components. There exists one big component. As one can see, there is a big part of the network that belongs to WG. It contains more classical graph theory papers. The GD dominated part is densely connected. In this part one can find a smaller fraction of WG papers.

GD WG citation networg

GD WG citation network.


WADS (yellow) and SWAT (red) are biannual alternating conferences on algorithms. Maybe the biggest difference is that SWAT takes place in Scandinavia, whereas WADS takes place in North America. Beside of this, both conferences have more or less the same scope. This reflects well in the citation network. It really gives the impression that both conferences are supported by the same community. The network itself is not very densely connected - it has several smaller connected components.

WADS SWAT citation network.

WADS SWAT citation network.


ICALP (yellow) and ESA (red) are two of the more general Europe-based conferences for Theoretical Computer Science. ICALP has a more general scope since it also covers explicitly formal languages and programming, whereas ESA has it focus on algorithms. The citation graph has two large components. One component is ICALP-only. The largest component is well-mixed, though some smaller parts are dominated by either ICALP or ESA.

ICALP ESA citation network.

ICALP ESA citation network.

Any ideas which other Springer conferences would be of interest?

One thought on “Citation graphs of interconnected communities

  1. Permalink  ⋅ Reply

    Dynamic Citation Graph - on graphs

    January 27, 2015 at 9:09am

    […] a year ago I had a post about citation networks of interconnected communities. Simply speaking, I displayed a citation network (which paper cites which) of two related […]

Leave a Reply