Current Events: Twitter - gt-big-data/wiki GitHub Wiki


Can we cluster tweets to find current events? Are some accounts more responsible for starting these clusters of tweets on a recent event? How do the clusters on recent events spread (virality)? Measure importance/relevance of information through virality?

Examples would be how does "The princess had a baby?" spread.


Name: Saman Shareghi, Utkarsh Garg, Philippe Laban, Patrick Violette


Some information about clustering analysis in the context of short messages:

github with sample code for streaming from twitter:

twitter streaming api (filter):

python natural language toolkit: