Tweet manual annotation and model scores: "Multifaceted polarisation and information reliability in climate change discussions on social media platforms"

CLINT Data

This repository contains the Twitter data that was annotated used in the project CLINT and the paper  "Multifaceted polarisation and information reliability in climate change discussions on social media platforms" .

The data includes tweets from four different datasets: COP27, IPCC, Climate Refugees and Doñana. Each dataset is represented by 80 tweets.The information included in the dataset is the following:

network_id – ID of the network the tweet belongs to.

community_user – Community of the source user.

community_referenced – Community of target user.

id – Unique tweet identifier.

original_tweet – Text of the tweet in its original language.

translated_text – Translated text of the tweet.

model_score – Score assigned by the predictive model.

negativity_score_reviewer1 – Negativity score from reviewer 1.

negativity_score_reviewer2 – Negativity score from reviewer 2.

negativity_score_reviewer3 – Negativity score from reviewer 3