This repo takes data from two excellent NYT articles and puts it in a more structured form. The articles in question are The People, Places and Things Trump Has Praised on Twitter: A Complete List and The 459 People, Places and Things Donald Trump Has Insulted on Twitter: A Complete List
Each of these articles catalouges Trumps tweets and highlights phrases that were an insult / compliment.
The final data (after our proccessing) looks like this
The data, after our modifications is in ny_times_data.csv in this repo. Just glance at the notebook to see what we did.
At LightTag we build tools to label text and we couldn't miss an opportunity to promote a wonderful labeled data set. The staf at NYT have hand labeled Trump tweets for (we think) a few years and built quite a wonderful dataset. This repo contains their original annotations as well as our code for making them a a little richer
We're using this data to jump start a wider scale, public labeling project where we'll be labeling trump tweets.
We'll be using these annotations to pre-populate a model, which will provide suggestions to our community of annotators.
We're putting this out there with the goal of building a community of "trump annotators", hopefully from all sides of the spectrum. Since this is a community effort, we'll be releasing all annotations that come in. You'll find the latest version once a week here. Anyone annotating can download the latest data live
Glad you asked. Just go to our demo page, signup and start labeling