Table 6.

Europe PMC dictionary-based entity annotation follows a sequential manner to annotate the entities.

	human annotation
		Gene/Protein	Disease	Organism
Europe PMC Annotation	Gene/Protein	—	324	113
	Disease	47	—	18
	Organism	19	110	—

For example, we apply the Gene/Proteins dictionary before the Disease dictionary, making the Gene/Protein terms unavailable for the disease tagger. We minimise the false positive identifications through this approach. This table shows the number of wrong entity type assignments by the Europe PMC approach corrected by the manual annotators. Europe PMC misses a small percentage of the Disease and Organism entities due to the sequential approach. We are showing Europe PMC annotation in the rows and the manually corrected ones in the columns.