论文信息 - A biclustering-based approach to filter dishonest advisors in multi-criteria e-marketplaces

A biclustering-based approach to filter dishonest advisors in multi-criteria e-marketplaces

In this paper, we propose a biclustering-based approach to identify dishonest advisors (who provide misleading opinions about sellers), while evaluating seller trustworthiness on multiple criteria. It considers correlation between advisors’ ratings to various criteria and trust transitivity to accurately filter the dishonest advisors. Evaluation results demonstrate the robustness of our approach against various types of unfair rating attacks.

Jie Zhang | Siwei Jiang | Athirai Aravazhi Irissappane | Jie Zhang | Siwei Jiang

[1] Ümit V. Çatalyürek,et al. A Biclustering Method to Discover Co-regulated Genes Using Diverse Gene Expression Datasets , 2009, BICoB.

[2] Chunyan Miao,et al. iCLUB: an integrated clustering-based approach to improve the robustness of reputation systems , 2011, AAMAS.

[3] Jie Zhang,et al. A Framework to Choose Trust Models for Different E-Marketplace Environments , 2013, IJCAI.

[4] Jie Zhang,et al. Evaluating the trustworthiness of advice about seller agents in e-marketplaces: A personalized approach , 2008, Electron. Commer. Res. Appl..

[5] Jie Zhang,et al. An evolutionary model for constructing robust trust networks , 2013, AAMAS.