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

Table 7

5-classification experimental results.

Model\Index	Acc	P	R	F-score
CNN	0.6885 ± 0.1145	0.6956 ± 0.1087	0.6781 ± 0.1065	0.6644 ± 0.1432
RNN	0.7252 ± 0.1010	0.7426 ± 0.1153	0.7176 ± 0.1354	0.7078 ± 0.1312
LSTM	0.7146 ± 0.1235	0.7347 ± 0.1154	0.7255 ± 0.1251	0.7118 ± 0.1165
Reference [31]	0.7487 ± 0.1005	0.7665 ± 0.1087	0.7533 ± 0.1214	0.7402 ± 0.1109
Reference [32]	0.7337 ± 0.1001	0.7640 ± 0.0978	0.7543 ± 0.0899	0.7311 ± 0.0792
Proposed	0.9004 ± 0.0743	0.9171 ± 0.0742	0.9014 ± 0.0735	0.9001 ± 0.0567