Fig. 2

Chemical entity mention (CEM) occurrence distribution for the ~87k CEMs found by ChemDataExtractor v2.1 when 1200 RSC papers of TADF research were extracted using the built-in ‘Compound’ model. The x-axis represents the number of papers in which a specific CEM occurs, while the y-axis shows the number of CEMs with a given occurrence count in log-scale. There is a highly linear region in the distribution as plotted in the inner box, implying that the number of CEMs drops exponentially with the occurrence count in this intermediate regime.