[Paper Review] A Dynamic Model for Sharing Reputation of Sellers among Buyers for Enhancing Trust in Agent Mediated e-market

Vibha Gaur, Neeraj Kumar Sharma+1 more|arXiv (Cornell University)|Jan 18, 2012

Access Control and Trust17 references5 citations

TL;DR

This paper proposes a dynamic reputation model for agent-mediated e-markets that filters dishonest advisor opinions and dynamically weights honest advisors' feedback based on transaction value and consistency. By monotonically increasing the reputation of reliable advisors in proportion to transaction value, the model enhances trust and resists manipulation in shared reputation systems.

ABSTRACT

Reputation systems aim to reduce the risk of loss due to untrustworthy participants. This loss is aggravated by dishonest advisors trying to pollute the e-market environment for their self-interest. A major task of a reputation system is to promote and encourage advisors who repeatedly respond with fair advice and to apply an opinion filtering or honesty checking mechanism to detect and resist dishonest advisors. This paper provides a dynamic approach to compute the aggregated shared reputation component by filtering out unfair advice and then generating the aggregated shared reputation value. The proposed approach is dynamic in nature as it is sensitive to the behaviour of advisors, value of the current transaction and encourages the cooperation among buyers as advisors. It provides incentive to honest advisors in lieu of repeated sharing of honest opinion by increasing the weight of their opinion and by making the increase in the reputation of honest advisors monotonically proportional to the value of a transaction.

Motivation & Objective

• To address the problem of dishonest advisors manipulating reputation systems in agent-mediated e-markets.
• To reduce the risk of loss from untrustworthy participants through a more robust reputation aggregation mechanism.
• To incentivize honest advisors by dynamically increasing their reputation weight based on transaction value and consistency.
• To develop a system that is sensitive to advisor behavior and current transaction context for improved trustworthiness.

Proposed method

• The model computes an aggregated shared reputation by filtering out unfair or dishonest advisor opinions.
• It dynamically adjusts the weight of each advisor’s opinion based on their past behavior and reliability.
• The reputation increase for honest advisors is monotonically proportional to the value of the current transaction.
• A feedback filtering mechanism detects and resists dishonest advisors by analyzing opinion consistency and deviation.
• The system encourages cooperation among buyers by rewarding consistent, truthful feedback with higher influence in reputation computation.
• The approach integrates multi-agent system principles with social network dynamics to maintain trust in decentralized e-marketplaces.

Experimental results

Research questions

• RQ1 How can a reputation system effectively detect and filter dishonest advisor opinions in agent-mediated e-markets?
• RQ2 What mechanisms can dynamically weight advisors' feedback based on their trustworthiness and transaction value?
• RQ3 How can the system incentivize long-term honest behavior among buyers acting as advisors?
• RQ4 To what extent does dynamic weighting improve the accuracy and resilience of shared reputation systems?
• RQ5 How does the model maintain trust under varying levels of advisor dishonesty and transaction value?

Key findings

• The model successfully filters out unfair or dishonest advisor opinions through a consistency-based detection mechanism.
• Honest advisors receive increased reputation weight in direct proportion to the value of the transaction, incentivizing truthful feedback.
• The dynamic weighting mechanism enhances the reliability of the aggregated reputation score by favoring consistent, high-quality contributors.
• The system demonstrates improved resistance to manipulation compared to static reputation models.
• The approach promotes cooperation among buyers by giving long-term honest contributors greater influence in reputation computation.
• The model maintains trustworthiness even when a subset of advisors act dishonestly, due to its adaptive filtering and weighting strategy.