Table 4 Experimental results for combined sessions- Rand (S1R + S2R) based on different pre-trained models
From: An evaluation of transfer learning models in EEG-based authentication
Model | Comparison of classification accuracy (Averaged% ± Standard Deviation) | ||||
|---|---|---|---|---|---|
Acc. | Pre. | Sens. | Spec. | F1 | |
GoogLeNet | 97.56 ± 0.27 | 97.76 ± 0.64 | 97.50 ± 0.27 | 99.91 ± 0.01 | 97.44 ± 0.20 |
Inception-V3 | 99.78 ± 0.08 | 99.79 ± 0.07 | 99.78 ± 0.07 | 99.98 ± 0.003 | 99.79 ± 0.07 |
ResNet-50 | 99.84 ± 0.08 | 99.84 ± 0.08 | 99.84 ± 0.07 | 99.98 ± 0.003 | 99.84 ± 0.08 |
ResNet-101 | 99.79 ± 0.08 | 99.79 ± 0.08 | 99.78 ± 0.07 | 99.98 ± 0.002 | 99.78 ± 0.08 |
EfficientNet-B0 | 99.73 ± 0.02 | 99.73 ± 0.01 | 99.73 ± 0.01 | 99.98 ± 0.002 | 99.73 ± 0.005 |
DenseNet-201 | 99.95 ± 0.02 | 99.95 ± 0.03 | 99.94 ± 0.01 | 99.99 ± 0.002 | 99.94 ± 0.002 |