Research Article
Detecting Fake Reviews with Generative Adversarial Networks for Mobile Social Networks
Table 3
Results on different model weights.
| | AUC | Precision | Recall | F1-score |
| 0 | 0.773 | 0.771 | 0.772 | 0.771 | 0.1 | 0.776 | 0.775 | 0.776 | 0.775 | 0.5 | 0.778 | 0.776 | 0.78 | 0.778 | 0.8 | 0.78 | 0.78 | 0.779 | 0.779 | 0.9 | 0.788 | 0.788 | 0.781 | 0.784 | 0.95 | 0.795 | 0.796 | 0.800 | 0.798 | 1 | 0.782 | 0.780 | 0.781 | 0.780 | 1.05 | 0.780 | 0.780 | 0.780 | 0.780 | 1.1 | 0.778 | 0.78 | 0.773 | 0.776 | 1.5 | 0.772 | 0.771 | 0.771 | 0.771 | 2 | 0.76 | 0.758 | 0.76 | 0.759 | 10 | 0.71 | 0.703 | 0.701 | 0.702 | 100 | 0.62 | 0.617 | 0.62 | 0.618 | 10000 | 0.433 | 0.435 | 0.433 | 0.434 |
|
|