/cuentalo

Primary LanguageJupyter Notebook

This repository holds the material for our research on the #cuéntalo phenomenon that happened during the end of April 2018 and beginning of May 2018 on twitter.

The full raw dataset can be obtained here https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/79KH5Z (a list of tweet IDs that need to be hydrated, https://github.com/DocNow/hydrator)

The aim of this project is to provide analytical insights on this topic, such as geolocalization of the tweets, dynamics of the hashtag and then a neural network classifier to identify the topics mentioned in the tweets, and the general profile of who wrote them.

Our study was presented in the LA-WEB 19 conference (part of The Web Conference), the article is here. The labeled data set we used for the study is here.