metadata
language:
- es
In May 2020, I programmed a bot that analyzed a live stream of Twitter headlines and retrieved only those with positive stories in the COVID-19 pandemic. I stopped the bot shortly after the advent of the vaccine. In this dataset, you will find some of the tweets that I used for training, labeled as pos_news (good or positive news), and not_pos_news (negative or neutral news).
This dataset would be useful to make similar news classifiers for other diseases, perhaps replacing some of its entities (e.g. replacing COVID disease mentions for other disease terms).