Figure 2

Characteristic words in each party’s tweets related to COVID-19 in the GloVe word embedding space. We detect over-represented words by calculating the log odds ratio of each word (see Sect. 2) and obtain the GloVe embeddings for each word. We use UMAP to reduce dimensionality and plot each word. Colors indicate topic labels that we assign. The Democratic party member’s tweets features more words about the pandemic and its disproportionate influences, while the Republican tweets features words about Trump and the White House as well as words about China