Abstract
Trust is essential to economic efficiency. Trading partners choose each other and make decisions based on how much they trust one another. The way to assess trust in e-commerce is different from those in brick and mortar businesses, as there are limited indicators available in online environments. One way is to deploy trust and reputation management systems that are based on collecting feedbacks about partners’ transactions. One of the problems within such systems is the presence of unfair ratings. In this paper, an innovative QADE trust model is presented, which assumes the existence of unfairly reported trust assessments. Subjective nature of trust is considered, where differently reported trust values do not necessarily mean false trust values but can also imply differences in dispositions to trust. The method to identify and filter out the presumably false values is defined. In our method, a trust evaluator finds other agents in society that are similar to him, taking into account pairwise similarity of trust values and similarity of agents’ general mindsets. In order to reduce the effect of unfair ratings, the values reported by the non-similar agents are excluded from the trust computation. Simulations have been used to compare the effectiveness of algorithms to decrease the effect of unfair ratings. The simulations have been carried out in environments with varying number of attackers and targeted agents, as well as with different kinds of attackers. The results showed significant improvements of our proposed method. On average 6% to 13% more unfair trust ratings have been detected by our method. Unfair rating effects on trust assessment were reduced with average improvements from 26% to 57% compared to the other most effective filtering methods by Whitby and Teacy.
Highlights
Internet has become an important business medium and there are growing number of participants engaging in electronic commerce
The QADE trust model, implies that trust is a personal and subjective phenomenon based on different factors, which have not been appropriately considered in existing methods for handling false trust values
The simulation results showed that the proposed QADE filter algorithm corrects agents’ trust attitude significantly more successful than the most representative endogenous filtering technique proposed by Whitby et al and the most representative exogenous filtering method proposed in TRAVOS model, with average improvement of 57% over Whitby and 26% over TRAVOS filter
Summary
Internet has become an important business medium and there are growing number of participants engaging in electronic commerce. The assessed trust value is misleading if false ratings from other entities are taken into the trust computation. This could result in wrong decisions with all the related consequences. An innovative trust model called QADE (Qualitative Assessment Dynamics – Extended) is presented, which provides a genuine solution for resolving unfair trust ratings while considering agents’ different trust tendencies. The QADE trust model, implies that trust is a personal and subjective phenomenon based on different factors, which have not been appropriately considered in existing methods for handling false trust values. A method for handling with unfair trust values is proposed in Section “Unfair ratings”, followed by the presentation and analysis of simulation results. Section “Related work” contains a review of related work and Section “Conclusions” completes the article
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
