Table 6 Comparison with the State-of-the-art on the DB1 and DB9 datasets.
Method | Year | DB1 | DB9 | ||||||
|---|---|---|---|---|---|---|---|---|---|
Accuracy | Precision | Recall | F1-Score | Accuracy | Precision | Recall | F1-score | ||
LS-SVM (IAV+MAV+RMS+WL)1 | 2021 | 85.19 | - | - | - | - | - | - | - |
LDA (IAV(or MAV)+ CC)1 | 2021 | 84.23 | - | - | - | - | - | - | - |
RF2 | 2019 | 75.32 | - | - | - | - | - | - | - |
RNN with weight loss40 | 2018 | 79.30 | - | - | - | - | - | - | - |
LSTM+MLP41 | 2018 | 75.45 | - | - | - | - | - | - | - |
Attention-based hybrid CNN-RNN11 | 2018 | 87.00 | - | - | - | - | - | - | - |
RCNN42 | 2022 | 87.34 | - | - | - | - | - | - | - |
CFF-RCNN42 | 2022 | 88.87 | - | - | - | - | - | - | - |
Proposed Model | - | 94.31 | 95.60 | 94.31 | 94.08 | 98.96 | 98.50 | 98.96 | 98.60 |
