Figure 9From: The growing amplification of social media: measuring temporal and social contagion dynamics for over 150 languages on Twitter for 2009–2020Language identification divergence. A normalized ratio difference \(\delta D_{\ell }\) (i.e., divergence) closer to zero implies strong agreement, whereby both classifiers captured approximately the same number of messages over the last decade. Grey bars indicate higher rate of messages captured by FastText-LID, whereas red bars highlight higher rate of messages captured by Twitter-LIDBack to article page