Table 6. Classification results of PPI predictions on the STRING database.

Combined-scores: PPIs that have combined scores are considered positive cases. Experimental-700: PPIs that have experimental scores over 700 are considered positive cases. Direct comparison: the results of embeddings using the same method (cbow) and same hyperparameters. Different embedding methods: the results of BioConceptVec (skip-gram), BioConceptVec (GloVe) and BioConceptVec (fastText). The highest results of each section are marked as bold.

	Combined-score dataset				Experimental-700 dataset
	Precision	Recall	F1	AUC	Precision	Recall	F1	AUC
Direct comparison
BioAvgWord (cbow)	0.8195	0.7935	0.8063	0.8941	0.8851	0.7422	0.8074	0.9123
Yu et al. (cbow)	0.8236	0.8017	0.8125	0.9029	0.9130	0.7686	0.8346	0.9283
BioConceptVec (cbow)	0.8304	0.8025	0.8162	0.9064	0.9476	0.7981	0.8664	0.9525
Different embedding methods
BioConcept (skip-gram)	0.8279	0.8097	0.8187	0.9074	0.9201	0.8525	0.8850	0.9522
BioConcept (GloVe)	0.8116	0.8102	0.8109	0.9004	0.8656	0.8289	0.8468	0.9218
BioConcept (fastText)	0.8324	0.8100	0.8210	0.9099	0.9076	0.8677	0.8872	0.9556