Towards a Fuzzy-Based Model for Human-like Multi-Agent Negotiation

Abstract

Intelligent agents are applied to automate the process of negotiation. Typical multi-agent negotiation strategies only allow the exchange of precise quantitative values as attributes. In the real-world applications, many of the negotiation issues such as qualitative concepts, cannot be expressed by exact numeric values.In this paper, we propose a new fuzzy-based model for negotiation. Our model leads to a human-like negotiation, and enables negotiation parties to act flexibly.

A Multi-Agent Environment for Negotiation

“…The other agent integrated into the negotiation system is the Fuzzy-based model (FBM) agent introduced in [29]. The Fuzzy-based agent is designed for negotiation where agents can exchange fuzzy proposals.…”

“…The other agent integrated into SAMIN is the FBM agent introduced in [29]. The FBM agent was tested in a setup where it has to negotiate against the Bayesian agent about a single issue defined on real values ranging from 10 to 30.…”

“…A number of the state-of-the-art agents have found a place in SAMIN: ABMP [17], Bayesian agent [14], Bayesian Tit-forTat [12], FBM [29], Trade-off agent [6], QO agent [24], Random Walker [11]. As they were developed by different teams, their design, architecture, and implementation varies.…”

Towards an Open Negotiation Architecture for Heterogeneous Agents

“…The code created for agents introducing new strategies in the literature typically has been developed with respect to one or a few specific domains, and to run against other agents implemented by the same team [14,7,17,19]. This is understandable, since the code is to provide evidence of the excellence of the strategy.…”

“…While the literature focuses on the development of ever more clever negotiation agents [14,6,7,17,19,11], the actual use of these agents in ecommerce applications is prohibited by two factors: the inflexibility of the agents and the lack of ecommerce applications that are open to such agents [13].…”