Learning Twitter User Sentiments on Climate Change with Limited Labeled Data

Allison Koenecke, Jordi Feliu-Fabà

Keywords: Climate Change, Twitter Data, Sentiment Analysis, Automated Labelling, Cohort Analysis

TL;DR: We train RNNs on famous Twitter users to determine whether the general Twitter population is more likely to believe in climate change after a natural disaster.

Abstract: While it is well-documented that climate change accepters and deniers have become increasingly polarized in the United States over time, there has been no large-scale examination of whether these individuals are prone to changing their opinions as a result of natural external occurrences. On the sub-population of Twitter users, we examine whether climate change sentiment changes in response to five separate natural disasters occurring in the U.S. in 2018. We begin by showing that tweets can be classified with over 75% accuracy as either accepting or denying climate change when using our methodology to compensate for limited labelled data; results are robust across several machine learning models and yield geographic-level results in line with prior research. We then apply RNNs to conduct a cohort-level analysis showing that the 2018 hurricanes yielded a statistically significant increase in average tweet sentiment affirming climate change. However, this effect does not hold for the 2018 blizzard and wildfires studied, implying that Twitter users' opinions on climate change are fairly ingrained on this subset of natural disasters.