The Alchemy of Data Mining

More than one month ago I started to datamine a small number of russian Twitter accounts. Somehow they are not directly connected with each other but are all attracted to some uncanny personalities and to the more or less various branches of russian state in a region of Moscow: the legislature, the executive, the judiciary, the media and the orthodox church, on a national and on a local scale. By looking on how the information was spread it became clear, that these are not russian citizens but rather a swarm of bots, accelerating the content of choice for the power organs. It’s complex network of spreading information without being identified as bots, rather as casual, patriotic and good looking humans mostly living around Moscow.
In this project I want to go deeper into this spreading model of information, by analyzing the hierarchy, the methods, the reactions and last but not least try to reconstruct the code for the found botnet and rebuild a possible model of this “pseudo-region”. It’s an interesting field of study because Twitter is not the most frequent platform for social interaction in Russia, most of the time the recognizable human users are not even aware of this content, so it’s easier to identify via a visual approach the structures of the information spread. I’m using only open-source software and the Twitter-API.