Computational Intelligence and Neuroscience / 2022 / Article / Tab 7 / Research Article
Stress Classification Using Brain Signals Based on LSTM Network Table 7 Comparison of stress classification accuracies.
Reference Number of subjects Number of electrodes Levels of stress Stimulus Classification method Accuracy (%) [31 ] 18 32 2 Emotional video clips Mean asymmetry scores — [37 ] 9 14 2 and 4 SCWT SVM 85.17 (2 classes) 67.06 (4 classes) [61 ] 12 3 3 SCWT SVM 72.3 [35 ] 7 14 3 Multitasking, SCWT SVM 77.53 Arithmetic calculations and memory [62 ] 28 4 2 and 3 Public speaking MLP 92.85 (2 classes) 64.28 (3 classes) [63 ] 10 1 2 SCWT SVM 97.6 [21 ] 9 14 2 High and low altitude FC-DNN 86.62 Construction site Proposed 35 4 2 Emotional video clips LSTM 21 87.22 Proposed 35 4 2 Emotional video clips LSTM 22 89.28 Proposed 35 4 2 Emotional video clips LSTM 23 90.33 Proposed 35 4 2 Emotional video clips LSTM 24 93.17
1 Result for 50–50 training-testing data, 2 result for 60–40 training-testing data, 3 result for 70–30 training-testing data, and 4 result for 10-fold cross validation.