A framework for trust modeling in multiagent electronic marketplaces with buying advisors to consider varying seller behavior and the limiting of seller bids

  • Authors:
  • Jie Zhang;Robin Cohen

  • Affiliations:
  • Nanyang Technological University, Singapore;University of Waterloo, Canada

  • Venue:
  • ACM Transactions on Intelligent Systems and Technology (TIST) - Special section on agent communication, trust in multiagent systems, intelligent tutoring and coaching systems
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this article, we present a framework of use in electronic marketplaces that allows buying agents to model the trustworthiness of selling agents in an effective way, making use of seller ratings provided by other buying agents known as advisors. The trustworthiness of the advisors is also modeled, using an approach that combines both personal and public knowledge and allows the relative weighting to be adjusted over time. Through a series of experiments that simulate e-marketplaces, including ones where sellers may vary their behavior over time, we are able to demonstrate that our proposed framework delivers effective seller recommendations to buyers, resulting in important buyer profit. We also propose limiting seller bids as a method for promoting seller honesty, thus facilitating successful selection of sellers by buyers, and demonstrate the value of this approach through experimental results. Overall, this research is focused on the technological aspects of electronic commerce and specifically on technology that would be used to manage trust.