Research Article
Abnormal Profiles Detection Based on Time Series and Target Item Analysis for Recommender Systems
Table 2
An example of rating matrix and attack profiles.
| | Item1 | Item2 | Item3 | Item4 | Item5 | ⋯ | Itemn |
| User1 | 5 | 2 | 3 | 0 | 0 | ⋯ | 5 | User2 | 2 | 0 | 4 | 1 | 2 | ⋯ | 3 | User3 | 4 | 2 | 3 | 0 | 5 | ⋯ | 0 | User4 | 0 | 3 | 0 | 3 | 4 | ⋯ | 3 | | | | | | | | | | 2 | 0 | 4 | 1 | 2 | ⋯ | 3 | Attacker1 | 2 | 1 | 0 | 0 | 5 | ⋯ | 4 | Attacker2 | 2 | 2 | 0 | 0 | 5 | ⋯ | 3 | Attacker3 | 1 | 2 | 0 | 0 | 5 | ⋯ | 2 | | | | | | | | | | 2 | 0 | 0 | 0 | 5 | ⋯ | 4 | Count (5) | 2 | 2 | 2 | 2 | 9 | ⋯ | 3 |
|
|