A Mental Workload Evaluation Model Based on Improved Multibranch LSTM Network with Attention Mechanism

Table 5

3-classification experimental results.

Model\Index	Acc	P	R	F-score
CNN	0.7731 ± 0.1065	0.7682 ± 0.1213	0.7843  0.1242	0.7654  0.1132
RNN	0.7845 ± 0.1023	0.7994 ± 0.1235	0.7653  0.1730	0.7598  0.1241
LSTM	0.8123 ± 0.1164	0.8345 ± 0.1163	0.7992  0.1253	0.8013  0.1151
Reference [31]	0.8632 ± 0.1010	0.8721 ± 0.1012	0.8543  0.1132	0.8646  0.1002
Reference [32]	0.8637 ± 0.0993	0.8606 ± 0.0412	0.8620  0.0317	0.8479  0.0924
Proposed	0.9514 ± 0.0362	0.9653 ± 0.0324	0.9247  0.0432	0.9365  0.0531