Table 4 Comparative outcome of SCGERS-STGCN model with existing approaches19,20,21,49,50,51.
Approaches | \(Accu_{y}\) | \(Prec_{n}\) | \(Reca_{l}\) | \(F1_{score}\) |
|---|---|---|---|---|
CNN-BiLSTM | 88.47 | 87.51 | 85.77 | 90.97 |
1D-CNN | 87.62 | 86.61 | 88.67 | 83.67 |
GPTDS | 87.95 | 86.85 | 85.09 | 90.33 |
PC-TTA | 86.86 | 86.01 | 87.89 | 83.14 |
Vis-Net | 84.24 | 92.11 | 81.87 | 84.25 |
ViT-B/16/SAM | 81.40 | 88.40 | 81.87 | 84.20 |
5-layer model | 87.70 | 83.70 | 86.37 | 85.15 |
ResNet-50 algorithm | 87.27 | 86.27 | 84.37 | 89.60 |
CNN classifier | 86.30 | 85.30 | 87.10 | 82.58 |
GoogleNet method | 83.56 | 91.56 | 81.27 | 83.48 |
Inception model | 91.60 | 90.60 | 86.77 | 88.97 |
CNN-raspberry | 94.97 | 91.97 | 81.02 | 81.27 |
SCGERS-STGCN | 98.53 | 93.79 | 89.49 | 91.13 |
