South African bot behaviour post the July 2018 Twitter account cull

Abstract

World wide, there is a concerning use of social media to sway public opinion through the use of disinformation campaigns. Elections, political decisions such as the Brexit referendum and topics such as vaccination have all been targets of what has come to be known as computational propaganda. Twitter maintainers finally acted on the problem and in July 2018 they suspended millions of fake accounts. Automated accounts, known as bots, are substantially easier to spot and analyse when they are newly created and in the “egg” stage. The culling of accounts therefore provided the perfect opportunity to look for common behaviour. This paper provides an account of the common behaviour seen on a sample of high profile South African focused Twitter accounts in the days following the action by Twitter maintainers to remove abusive accounts.