This dataset features the output of intermediate steps and the final output of the research that is described in the paper: F. Kunneman and A. Van den Bosch (2014), Event detection in Twitter: A machine-learning approach based on term pivoting, In: F. Grootjen, M. Otworowska, and J. Kwisthout (Eds.), Proceedings of the 26th Benelux Conference on Artificial Intelligence, pp. 65-72, http://hdl.handle.net/2066/132203 The paper describes an approach to extracting events (of all types) from Twitter by identifying words that are suddenly mentioned frequently in tweets, clustering similar terms together and training a model to distinguish significant from insignificant event clusters. The current dataset features the word counts that are used to find words with a sudden high frequency, the clusters of such words, and the classification decisions for event significance.   Tweets relate to the following period: June 22nd 2013 - August 22nd 2013.