Towards Autonomous and Automatic Evaluation and Negotiation in Agent-Mediated Internet Marketplaces

  • Authors:

  • Yan Wang;Kian-Lee Tan;Jian Ren

  • Affiliations:

  • Department of Computing, Macquarie University, Sydney, Australia 2109;Department of Computer Science, National University of Singapore, Singapore 117543;Department of Computer Science, National University of Singapore, Singapore 117543

  • Venue:
  • Electronic Commerce Research
  • Year:
  • 2005

Quantified Score

Hi-index 0.01

Visualization

Abstract

In an agent-mediated B2C Internet marketplace, agents, mobile or stationary, can act on behalf of customers to complete tasks involved in buying and selling products. In this paper, we first present a fuzzy-set based two-phase evaluation model allowing mobile agents to evaluate and filter online e-shops, and evaluate offers autonomously and automatically at the request from a consumer. The fuzzy evaluation criterion takes into account not only the attributes of offers, but also the reputation of corresponding e-shops. }Meanwhile, the two-phase operations can help reduce the scale of mobile agents dispatched to e-shops and thus reduce the network workload. Based on the same fuzzy evaluation criterion, an auction-like negotiation model is presented, where the consumer agent can autonomously determine the initial offer and negotiate with multiple shops simultaneously. The final best offer is determined not only by the attributes of the offers, but also the attributes of the e-shops as well as the preference the consumer specified. Finally, our models presented in this paper have been implemented in a prototype system where we conducted some experiments to evaluate their performance.