TwitCID: a Collection of Data Sets for Studies on Information Diffusion on Social Networks

The TwitCID collection consists of five Twitter datasets which were extracted from the 1 percent of tweets from Twitter API. 

The Firstweek and Secondweek data set were collected during the first week and second week of January 2017 while the Iphone, Gucci and Galaxy data sets were collected from 21 September 2015 to 31 May 2017 using the keywords “iphone”, “gucci” and “galaxys” respectively.  

We publish these datasets on behalf of our academic institution – IRIT, France and for the sole purpose of non-commercial research under the license CC BY-NC-SA (Attribution-NonCommercial-ShareAlike). In accordance with Twitter's Terms of Service, we only provide identifiers of tweets. In order to collect the actual tweets in JSON, you could use the script Collect_JSONtweets.py attached.

If you would like to use this collection, please cite our paper: 

Hoang, T. B. N., Mothe, J., & Baillon, M. (2019, September). TwitCID: a collection of data sets for studies on information diffusion on social networks. In International Conference of the Cross-Language Evaluation Forum for European Languages (pp. 88-100). Springer, Cham.