Table 3. Performance metrics for natural language processing and classification on the validation cohort across all outcomes for BOW with logistic regression and RNN with GloVe.

a) Average AUC metric across all five splits of the data. b) Sensitivity, Specificity, Accuracy and Precision for GloVe Models combined with RNN on the BMC Validation Cohort.

a)
	Stroke		Location		Acuity
BOW+Log.Reg	0.892 (0.875:0.91)		0.857 (0.845:0.869)		0.797 (0.768:0.828)
GloVe+RNN	0.920.908:0.932)		0.893.88:0.905)		0.925.906:0.946)
b)
	Sensitivity	Specificity		Accuracy		Precision
Stroke	0.915	0.752		0.828		0.764
MCA Location	0.898	0.7		0.862		0.932
Acuity	0.914	0.689		0.866		0.916