Machine Learning-Based Forecast of Hemorrhagic Stroke Healthcare Service Demand considering Air Pollution

Table 7

Statistics on the performance of warm season models without considering air pollution among the machine learning methods.

Model	M-AUC	M-Sens	M-Spec	SD-AUC	SD-Sens	SD-Spec
LR	0.6062–	0.4137–	0.8543–	0.1452+	0.3688+	0.0367+
RF	0.6504–	0.3750+	0.8571+	0.1011+	0.4361+	0.0444+
SVMLinear	0.6229–	0.2996+	0.8217+	0.1004+	0.2943+	0.0812–
KNN	0.6931+	0.3667+	0.8510–	0.1173+	0.4830+	0.0469+
XGBLinear	0.6783+	0.3600+	0.8590+	0.1415+	0.3719+	0.0428+
XGBTree	0.6453–	0.3333–	0.8422–	0.0881+	0.4157+	0.0258+

M-AUC, M-Sens, and M-Spec denote the average area under the curve (AUC), sensitivity, and specificity, respectively; SD-AUC, SD-Sens, and SD-Spec denote the standard deviation of the AUC, sensitivity, and specificity, respectively. LR, logistic regression; RF, random forest; SVMLinear, support-vector machines with linear kernel; KNN, k-nearest neighbor algorithm; XGBTree, extreme gradient boosting decision tree; XGBLinear, extreme gradient boosting linear model. “+” indicates that the corresponding value without considering air pollution is higher than that considering air pollution. “–” indicates that the corresponding value considering air pollution is higher than that without considering air pollution.