Table 3 Comparative assessment of the proposed framework upon different configurations
From: EEG-based image classification via a region-level stacked bi-directional deep learning framework
| Â | Framework | |||||
|---|---|---|---|---|---|---|
Configurations | 1 | 2 | 3 | 4 | 5 | 6 |
Lateralization | Region-level info. | Region-level info. | None | Region-level info. | Region-level info. | Region-level info. |
Feature encoder | Unidirectional LSTM | Unidirectional LSTM | BiLSTM | BiLSTM | BiLSTM | Stacked biLSTM |
Classifier | SoftMax | ICA+SVM | SoftMax | SoftMax | ICA+SVM | ICA+SVM |
Accuracy | 92.9% | 94.3% | 95.3% | 97.0% | 97.1% | 97.3% |