Twitter's Terms of Service restricts the distribution of tweet contents. To ensure research reproducibility, we publish our datasets in a manner similar to the TREC Microblog Track in which a list of tweet IDs is released instead. Through this list, tweet content can be crawled from the Twitter public API through the statuses/show/:id endpoint.
Tools to automatically retrieve tweet contents from their ids are not provided by the authors, but some have been developed within the social sensor project.
Note that some tweets might have been deleted or their users changed the privacy level.
For more information, contact: jc@ac.upc.edu