Table 3 Training and test MCE in the LH dataset across 10 runs. Results are expressed in terms of average (standard deviation)
From: Feature stability and setup minimization for EEG-EMG-enabled monitoring systems
Model | Training | Test | TPR | TNR | F1 |
|---|---|---|---|---|---|
FeSC-HS | 0.1320 (0.0040) | 0.1127 (0.0331) | 0.8279 (0.2517) | 0.8049 (0.2439) | 0.8174 (0.2489) |
FeSC-SD | 0.1742 (0.0047) | 0.1799 (0.0368) | 0.8254 (0.0483) | 0.8149 (0.0387) | 0.8208 (0.0375) |
FeSC-HD | 0.1823 (0.0040) | 0.1537 (0.0315) | 0.8433 (0.0412) | 0.8493 (0.0382) | 0.8457 (0.0324) |
FeSC-HSD-0.6 | 0.1717 (0.0049) | 0.1552 (0.0283) | 0.8299 (0.0452) | 0.8597 (0.0346) | 0.8421 (0.0301) |
FeSC-HSD-0.9 | 0.1757 (0.0044) | 0.1642 (0.0382) | 0.6797 (0.3515) | 0.6725 (0.3478) | 0.6765 (0.3497) |
pFSFS | 0.0308 (0.0089) | 0.0604 (0.0181) | 0.9090 (0.0355) | 0.9701 (0.0157) | 0.9374 (0.0196) |
LASSO | 0(0) | 0.2007 (0.0308) | 0.6940 (0.0577) | 0.6940 (0.0577) | 0.5804 (0.0205) |