We have collected a large set of tweets observing the German federal election in 2017. Tweets originate from candidates to the German parliament, from different media accounts and from people who tweeted using a prespecified set of hashtags. Some of the account holders are described by metadata specifying, e.g. their party or their candidate status. From a political science perspective it would be interesting to find out who is able to influence other people and media accounts in picking up their stories, supporting them or attacking them. The notion of “picking up a story” needs to be modeled by methods from information retrieval (e.g. keyword overlap). Based on such story similarities, the task is to be modelled by learning association rules in a time series of events. Temporal association rules consist of premises describing that an events happens and a conclusion that describes that a consequence results over time from the premises.
- Association rules in general: Kotsiantis, Sotiris, and Dimitris Kanellopoulos. "Association rules mining: A recent overview." GESTS International Transactions on Computer Science and Engineering 32.1 (2006): 71-82. http://www.csis.pace.edu/~ctappert/dps/d861-13/session2-p1.pdf
- Temporal association rules: