This just replicates some other Python based analyses on the GDELT data, for example the one by David Masad here, but using %sql magic for IPython and pandas.
Eventually there will be a MySQL front-end provided on the data website, so that you don't have to download all of the data in the first place.
Download the MySQL table here (687 MB compressed, ca. 5.5 GB uncompressed). You need MySQL server running and Python's MySQLdb package to use the IPython notebook. Extract the file Events.sql.gz and import the table:
$ mysql -u yourusername -p -D yourdbname < Events.sql
Here's the nbviewer link to the notebooks: