The Project contains the Data and Codes used in the article:
"Why polls fail to predict elections". Authors: Zhenkun Zhou, Matteo Serafino, Luciano Cohan, Guido Caldarelli and Hernan A. Makse.
The Data folder contains raw_tweet_ids.zip, a list of the ID of tweets, retweets, and quotes for the 2019 Argentinian election used in the paper.
Softwares such as [hydrator](https://github.com/DocNow/hydrator) and [tweepy](https://www.tweepy.org/) can be used to “rehydrate” the tweet_IDs,
i.e. download the full tweet objects using the tweet_IDs.
The Codes folder we include all the python scripts used to produce the results in the paper.