Table 4.

The improvement of $O{A}_{all}$ achieved by the proposed algorithm, compared with the performance of the individual RNN, GRU, and LSTM algorithms.

	Driver Stress Prediction			Driver Drowsiness Prediction
	RNN	GRU	LSTM	RNN	GRU	LSTM
Improvement (%)	11.2	13.6	12.3	10.2	12.2	11.2