Effectiveness of Reverse Edges and Uncertainty in PIN-TRUST for Trust Prediction
- Title
- Effectiveness of Reverse Edges and Uncertainty in PIN-TRUST for Trust Prediction
- Author
- 김상욱
- Keywords
- PIN-TRUST; trust prediction; graph mining
- Issue Date
- 2016-10
- Publisher
- EDB 2016
- Citation
- Proceedings of the Sixth International Conference on Emerging Databases: Technologies, Applications, and Theory, Page. 81-85
- Abstract
- Recently, PIN-TRUST, a method to predict future trust relationships between users is proposed. PIN-TRUST out-performs existing trust prediction methods by exploiting all types of interactions between users and the reciprocation of ones. In this paper, we validate whether its consideration on the reciprocation of interactions is really effective in trust prediction. Furthermore, we consider a new concept, the "uncertainty" of untrustworthy users that is devised to reflect the difficulty on modeling the activities of untrustworthy users in PIN-TRUST. Then, we also validate the effectiveness this uncertainty concepts. Through the validation, we reveal that the consideration of the reciprocation of interactions is effective for trust prediction with PIN-TRUST, and it is necessary to regard the uncertainty of untrustworthy users same as that of other users.
- URI
- https://dl.acm.org/citation.cfm?doid=3007818.3007827https://repository.hanyang.ac.kr/handle/20.500.11754/81352
- ISBN
- 978-1-4503-4754-9
- DOI
- 10.1145/3007818.3007827
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE(컴퓨터소프트웨어학부) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML