On December 9th till December 11th the Utrecht Master programme New Media and Digital Culture had a team that participated in the R-Shief #OccupyData hackathon. The team consisted of Ryanne Turenhout, Ruben Hazelaar, Thomas Boeschoten and Mirko Schäfer. The hackathon was a three day event in which teams around the world (MIT, Cambridge, Hong Kong, and Los Angeles among others) participated in a hackathon to critically analyse more than 5 million Occupy tweets.
Above you see a screenshot of 14 @reply hashtag networks visualised in Gephi. Each colour represents a different hashtag @reply network. The hashtags are Maspero, OccupyAmsterdam, OccupyAtlanta, OccupyBoston, OccupyCal, OccupyCanada, OccupyChi, OccupyChicago, OccupyDallas, OccupyDavis, OccupyLSX, OccupySeattle, OccupySF, and OccupyWallst. This is still experimental analysis because some users are in different hashtag networks, only in the visualisation they are displayed in one hashtag network (indicated by colour).
Other results from the New Media & Digital Culture Utrecht hackathon team are going to be published soon this is only part of it. Our focus has been on the #OccupyAmsterdam dataset and we are still doing research on this dataset. An article about the OccupyAmsterdam dataset will be published later today.
Scripts used to create a GML file from data in a mysql database or extract data from csv files (the cleaned up versions that we have used).
The preliminary #OccupyAmsterdam network graph
The visualisation of 14 hashtag networks (pdf, legend and disclaimer)
The GML files merged into 1 file
Timeline of the number of tweets per day, per hashtag dataset
A graph with the most retweeted people in the #OccupyAmsterdam dataset
