Event Graph Layout

Introduction

I made a layout plugin for Gephi that can be used for the visualization of event graphs, where events are represented by nodes and the relationships between events by arcs/edges.

Since Gephi was updated to version 0.9, the plugin can only be found from within the Gephi program itself, as far as I know. However, there should be a new online market place at some point. To download the latest version of the plugin (for Gephi 0.9 and higher), run Gephi, enter the ‘Tools’ menu in the menu bar at the top, and click ‘Plugins’. The Event Graph Layout plugin should be in the list of available plugins, and can be installed from there. An older version of the plugin (for Gephi 0.8.2) is still available from the original Gephi Market Place if required. I recommend using the new version, but if you have good reasons for running Gephi 0.8.2, instead of the latest version, then the 0.8.2. version of the layout can be downloaded and installed manually (download the .nbm file, go to the plugins menu within Gephi [see instructions above], select ‘Downloaded’, and then select the downloaded .nbm file).

If you would like to cite this software, you can use the following format:

Spekkink, W. (2014). Event Graph Layout [software]. Available from: https://marketplace.gephi.org/plugin/event-graph-layout/

See this publication for an example of the tool in use.

Useful advice on the order variable

To some users it may not be immediately clear how to choose a suitable order variable. Knowing some things about the Gephi plot space come in useful here. I didn’t check whether the information provide below is fully correct. I more or less deduced it from my own experience with Gephi and the Event Graph Layout. The Gephi plot space appears to you as a space that is simply blank, but it does have two axes with positive as well as negative values.

PlotSpace2

It is possible to hit the edges of the plot space. So, I advice against using very large values for the order variable. The smaller the value you use, the more you will be able to fit in the plotting space. Also, if you also use negative values, you will be able to fit more in the plotting space, because with positive values you only utilize one half of the available space (this is one of the things I am not absolutely sure about, but using negative values does work).

If you want to make the order variable that you assign to each event proportional to the actual date that the event occurred, then you can use the following trick (or something similar). Let’s assume that you want the order variable to be proportional to the year and month in which the event occurred. In your nodes list, you can then report the year and month of occurrence in separate columns. You can then write a formula that converts these values into a suitable order variable (see table below for the formula). You should take the earliest reported year as the base year. For each event you subtract the base year from the year reported for that event. In the example below, for Event4 the result is 2003-2000 = 3. Then you add the number of months times 0.083 (which is roughly 1/12) to that number. For Event4 we can complete the formula: 2003-2000+(7*0.083) = 3.581. You only have to write this formula once in Excel and you can then copy it to the other rows.

Table

As I suggested, the order variable will now be proportional to the data that the events occurred (accurate to month and year). This makes it relatively easy to add a timeline to your plot after you exported it from Gephi. For example, in the picture below I simply imported the picture into Powerpoint, and I added a table that reports the years (the advantage of using a table is that each column will have the same size in its default make up). Then I stretched the table to make sure that it roughly matches the distribution of events.

Example