Graph-Based Link Prediction between Human Phenotypes and Genes

Table 2

Micro, macro, and weighted metrics.

Model	Metric	Micro	Macro	Weighted	AUROC	AUCPR
Logistic regression	Precision	0.70	0.55	0.89	0.70	0.23
	Recall	0.70	0.66	0.70
	F1 score	0.70	0.52	0.77
Random forest	Precision	0.94	0.97	0.94	0.75	0.46
	Recall	0.94	0.59	0.94
	F1 score	0.94	0.63	0.92
Neural network	Precision	0.87	0.66	0.93	0.83	0.28
	Recall	0.87	0.83	0.87
	F1 score	0.87	0.70	0.89
XGBoost	Precision	0.95	0.97	0.95	0.81	0.60
	Recall	0.95	0.67	0.95
	F1 score	0.95	0.74	0.94
LightGBM	Precision	0.84	0.64	0.93	0.90	0.78
	Recall	0.84	0.83	0.84
	F1 score	0.84	0.67	0.87