Table 6.

Popular evaluation metrics used for PHI prediction.

Metric	Formula	References
Accuracy	$\frac{TP+TN}{TP+FP+TN+FN}$	Cui et al., 2012
Specificity	$\frac{TN}{TN+FP}$	Cui et al., 2012
Sensitivity (Recall)	$\frac{TP}{TP+FN}$	Dyer et al., 2011; Cui et al., 2012
Precision	$\frac{TP}{TP+FP}$	Dyer et al., 2011
F1 score	$\frac{2*Precision*Recall}{Precision+Recall}$	Kshirsagar et al., 2012, 2013b; Mei, 2013; Coelho et al., 2014
AUC	The area under the ROC curve	Davis et al., 2007; Mei, 2013; Coelho et al., 2014

TP, True Positive; TN, True Negative; FP, False Positive; FN, False Negative.