Text mining attitudes towards climate change: Emotion and sentiment analysis of the Twitter corpus

Abstract

Media such as Twitter has become a platform for contemporary Americans to express their opinions and allow for reaction to public opinion. Climate change is a topic of ongoing social concern. Machine-automated processing facilitates big data analysis and is suitable for analyzing a large corpus of tweets. This paper uses R tools to conduct sentiment calculation and emotion analysis on the tweet corpus from 2015 to 2018 to present the overall tendency of citizens’ attitudes toward climate change topics. The keyword analysis finds that people focus on the message’s source; CLIMATE CHANGE and GLOBAL WARMING make an association. Supporters express FEAR and SURPRISE about extreme weather and opponents’ behavior, while opponents show ANGER, DISGUST and SADNESS about politicians manufacturing climate change stories about which they have no real feelings. This study also reveals that the automatic annotation tools are still inadequate, with limited emotion lexicon and identification of negation and sarcasm.