Research Article
An End-to-End Rumor Detection Model Based on Feature Aggregation
Table 4
Performance of proposed propagation pattern model (R: rumor; N: nonrumor).
| Method | Label | Accuracy | Precision | Recall | |
| DT-Rank | R | 0.732 | 0.738 | 0.715 | 0.726 | N | 0.726 | 0.749 | 0.737 |
| SVM-RBF | R | 0.818 | 0.822 | 0.812 | 0.817 | N | 0.815 | 0.824 | 0.819 |
| DTC | R | 0.831 | 0.847 | 0.815 | 0.831 | N | 0.815 | 0.847 | 0.830 |
| RFC | R | 0.849 | 0.786 | 0.959 | 0.864 | N | 0.947 | 0.739 | 0.830 |
| SVM-TS | R | 0.857 | 0.839 | 0.885 | 0.861 | N | 0.878 | 0.830 | 0.857 |
| CNN-Top | R | 0.874 | 0.878 | 0.874 | 0.876 | N | 0.869 | 0.873 | 0.871 |
| RNN-Top | R | 0.868 | 0.853 | 0.897 | 0.874 | N | 0.886 | 0.838 | 0.861 |
|
|