Spot A Bot

Bots on Twitter are used to spread false information, scam people, or influence politics. They often do not act alone, but in a coordinated manner to steer entire trends or even get them rolling. Which trends are organic and which are artificially grown can be difficult to detect when scrolling the Twitter timeline. Our AI solution is designed to detect bots across entire trends and make the extent of their activity visible to everyone.

To provide an overview of the amount of automatically generated content on Twitter, we designed a machine learning model that would be trained and tested with datasets of Twitter accounts and examine current Twitter trends on a daily basis. We wanted to make the results freely available on the web and present them in a comprehensible and transparent way.

Spot A Bot is a platform that allows visitors to check daily as well as past Twitter trends for bots on their own. The machine learning model provides an estimation of the bot share and gives insight into the analyzed tweets and accounts. With our model, we have already been able to uncover Chinese botnets that have attempted to hijack and suppress trends related to the protests there.