Description

Despite overwhelming scientific consensus regarding climate change, the basis for climate change still remains a widely debated and political issue. Our study investigates one particular dataset of over 600M Twitter posts in English to understand how the climate change sentiment has changed over time. We train a series of word2vec models, one per month, to examine how words used in the context of climate change have changed over time. We also fine-tune a BERT model to perform sentiment classification on tweets.