Name
Abstract | ||
|---|---|---|
Due to the open nature of collaborative recommender systems, they can not effectively prevent malicious users from injecting fake profile data into the ratings database, which can significantly bias the system's output. With this problem in mind, in this paper we introduce the social trust of the users into the recommender system and build the trust relation between them. The values of trust among users are adjusted by using the reinforcement learning algorithm. On the basis of this, a user trust-based collaborative filtering recommendation algorithm is proposed. It uses the combined similarity to generate recommendation, which considers not only the similarity between user profiles but user trust as well. Experimental results show that the proposed algorithm outperforms the traditional user-based and item-based collaborative filtering algorithm in recommendation accuracy, especially in the face of malicious profile injection attacks. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1007/978-3-642-11145-7_32 | ICICS |
Keywords | Field | DocType |
recommendation algorithm,collaborative recommender system,user profile,malicious user,item-based collaborative,recommendation accuracy,user trust,social trust,proposed algorithm,trust relation,collaborative filtering,recommender system,reinforcement learning | Recommender system,Injection attacks,Collaborative filtering,Computer science,Algorithm,Reinforcement learning algorithm,Social trust,Reinforcement learning | Conference |
Volume | ISSN | ISBN |
5927 | 0302-9743 | 3-642-11144-0 |
Citations | PageRank | References |
9 | 0.47 | 12 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Fuzhi Zhang | 1 | 23 | 6.53 |
| Long Bai | 2 | 20 | 3.82 |
| Feng Gao | 3 | 9 | 0.47 |