tlax28/african-violence

Code to create networks based on the ACLED data. # african-violence Code to create networks based on the ACLED data. Codebook: acleddata.com If you get an 'isinstance' error, replace the existing gml.py with the gml.py from Justin Ruths in the 'site-packages/zen/io' directory. (email dated 9 April) ##THINGS WE CAN DO: - For the two graphs: Event count network, Fatalities network - Find degree distribution, does it follow a power law? - Diameter for the giant component - Degree / Eigenvector centrality (parties most involved in conflict) - Find modularity value for the different classes (e.g. civilians vs. non-civilians, Govt vs rebels) ## Actor Categories All actors fall into 1 of 8 categories: Government or mutinous force = 1 Rebel force = 2 Political militia = 3 Ethnic militia = 4 Rioters = 5 Protesters = 6 Civilians = 7 Outside/external force (e.g. UN) = 8