Research Article
Semisupervised Graph Neural Networks for Traffic Classification in Edge Networks
Table 2
Related works based on deep learning models (multimodal architectures).
| | Research | Input data | TO | DL classifier | MM | MT | SSR | TPS | O |
| | Aceto et al. [17], Elsevier ComNet | (1) First 576 bytes of L4 payload | B | 1D-CNN&BiGRU | Y | N | — | Y | P | | (2) 4 fields for the first 12 packets |
| | Aceto et al. [15], JNCA | (1) First 784 bytes of L4 payload | B | 1D-CNN&BiGRU | Y | Y | Y | Y | Y | | (2) 4 fields for the first 32 packets |
| | Nascita et al. [16], TNSM | (1) First 576 bytes of L4 payload | B | 1D-CNN&BiGRU | Y | N | — | Y | Y | | (2) 4 fields for the first 12 packets |
|
|