Table 3:

The mean (standard deviation) of performance metrics of the support vector machine (SVM) algorithm adopted for identification of sepsis patients in the four data sets of the study. The features used are statistics of individual responses (Ind) and their combinations with age, gender, and medical history of patients (${\text{Ind}}^{+}$).

Gap (hours)	Features	Accuracy	Recall	Precision
3	$\text{Ind}$	0.72 (0.02)	0.69 (0.03)	0.70 (0.03)
	${\text{Ind}}^{+}$	0.76 (0.02)	0.77 (0.02)	0.72 (0.02)
6	$\text{Ind}$	0.71 (0.02)	0.68 (0.03)	0.69 (0.03)
	${\text{Ind}}^{+}$	0.75 (0.02)	0.76 (0.02)	0.71 (0.03)
12	$\text{Ind}$	0.71 (0.02)	0.69 (0.03)	0.70 (0.03)
	${\text{Ind}}^{+}$	0.75 (0.02)	0.76 (0.02)	0.72 (0.03)
24	$\text{Ind}$	0.68 (0.02)	0.65 (0.03)	0.68 (0.04)
	${\text{Ind}}^{+}$	0.73 (0.02)	0.74 (0.04)	0.71 (0.03)