Table 3.

Summary of the maximum and minimum $O{A}_{all}$, $O{A}_{class1}$, $O{A}_{class2}$, and $O{A}_{class3}$ for the driver stress prediction and driver drowsiness prediction models based on boosting learning of multiple RNNs, GRUs, and LSTMs algorithms.

Overall Accuracy	Driver Stress Prediction						Driver Drowsiness Prediction
	Minimum			Maximum			Minimum			Maximum
	RNNs	GRUs	LSTMs	RNNs	GRUs	LSTMs	RNNs	GRUs	LSTMs	RNNs	GRUs	LSTMs
$O{A}_{all}$ (%)	63.2	65.5	70.1	87.1	82.6	85.2	65.3	67.3	69.5	88.1	86.5	87.2
$O{A}_{class1}$ (%)	63.8	66	70.3	87.5	83.3	85.5	66.1	68	71	88.8	87	87.7
$O{A}_{class2}$ (%)	63.5	65.7	70.5	87.3	82.6	85.6	65	67	69	87.7	86.2	87
$O{A}_{class3}$ (%)	61.2	63.6	68.6	85.4	80.8	83.6	62.8	65.4	68.1	85.9	84.8	85.4