Research Article
A Shilling Group Detection Framework Based on Deep Learning Techniques
Table 1
Summary of threat models.
| Threat model | IS | IF | IT | Items | Ratings | Items | Ratings | Rating |
| GSAGenl Ran | Null | Item is randomly chosen and rated by at most two attackers | System mean | rmax or rmin |
| GSAGenl Avg | Null | Item is randomly chosen and rated by at most two attackers | Item mean | rmax or rmin |
| GSAGenl AoP | Null | Item is chosen randomly from x% popular items and rated by at most two attackers | Item mean | rmax or rmin |
| GSAGenl PIA | Items are randomly chosen from power items and rated by less than three attackers | Item mean | Null | rmax or rmin |
| GSAGenl GOAT | Items are randomly chosen from popular items rated by active users and rated by less than three attackers | Learned by GAN | Items are randomly chosen from popular items rated by active users and rated by less than three attackers | Learned by GAN | rmaxā1 |
|
|