Towards trust inference from bipartite social networks

O'Doherty, Daire; Jouili, Salim; Roy, Peter Van

doi:10.1145/2304536.2304539

Cited by 11 publications

(9 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CTRM algorithm introduced trust mechanism into collaborative filtering and effectively improves the cold start problem. Therefore, compared to literature [10], the CTRM algorithm improved the accuracy of recommendation to a certain extent. …”

Section: Bmentioning

confidence: 90%

“…However, PeopleRank was better than collaborative filtering algorithms for cold start user. Compared with PeopleRank and collaborative filtering, the literature [10] also increased the recommended effective to some extent. CTRM algorithm introduced trust mechanism into collaborative filtering and effectively improves the cold start problem.…”

Section: Bmentioning

confidence: 99%

“…Reputation value is aggregated by BETA density function in the Best algorithm in the literature [9]. A prediction technique of trust relationship for classification was put forward based on user similarity and globe reputation value according to sociological theory in the literature [10]. Malicious nodes can easily elevate its own global reputation value via the construction of anonymous neighbor nodes.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Modeling for Comment Trust Recommendation Based on Collaborative Filtering

Liao¹,

Wang²,

Liu³

et al. 2015

Proceedings of the International Conference on Logistics, Engineering, Management and Computer Science

View full text Add to dashboard Cite

Abstract-Asymmetry in the parties of the transaction leads to uncertainty in the transaction. Trust problem has been one of the bottlenecks restricting the development of ecommerce. For e-commerce product review issues, comment trust recommendation model was proposed based on comment credibility degree and user similarity, which combined with social networking trust mechanism and collaborative filtering to offer users with a more personalized trust recommendations. The experimental results demonstrate that the model can effectively improve the recommendation accuracy and solve cold startup problems in collaborative filtering.

show abstract

Section: Bmentioning

confidence: 90%

Section: Bmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling for Comment Trust Recommendation Based on Collaborative Filtering

Liao¹,

Wang²,

Liu³

et al. 2015

Proceedings of the International Conference on Logistics, Engineering, Management and Computer Science

View full text Add to dashboard Cite

show abstract

“…The main contribution of our work is summarized as follows: -The social network is modeled as a directed graph with three types of relationships: the friendship, family and professional relationship. What is not considered in the previous work on trust [11] [13].…”

Section: Introductionmentioning

confidence: 99%

Inferring trust relationships in the social network: Evidence theory approach

Mekouar

Ibrahimi

Bouyakhf

2014

2014 International Wireless Communications and Mobile Computing Conference (IWCMC)

View full text Add to dashboard Cite

The trust is one of the crucial and important factors for decision making in online social network. In this paper, we compute explicitly the level of trust, distrust and uncertainty between two users using social interactions. The study of social interactions is fundamental to understanding the dynamics of relationships between users and their behaviors in the social network. We use later this formulation to infer trust between a user and a stranger who has no direct interaction with this latter. Finally, we use the Dempster-Shafer theory to combine the different views of different users to infer trust between a user and a stranger who is connected through neighbors by introducing the friendship, family and professional relationship. The numerical study of the proposed model shows its acceptable performance.

show abstract

“…As such, constructing personal recommendation systems that take on-line social interactions into consideration has been a hot research topic [3,16,17,32,29,28,35]. In addition to the user-item network that collects historical interactions between users and items, there is another social network (sometimes called trust network) that characterizes the interactions among users.…”

Section: Introductionmentioning

confidence: 99%