Agent Multi-issue Negotiation with Cases

Seller agents usually have more domain knowledge and more experiences which would help them get better negotiation result. This inspires us to investigate negotiation using case-based method in order to remember previous cases and reuse information and knowledge of that situation for new offers. Moreover, it can overcome two shortcomings in agent multi-issue negotiation researches. One is that there is a lack of the use of agent experiences. The other is that there is a lack of agent optimal offer strategy. To this end, a case-based agent multi-issue negotiation model is proposed. The emphasis of the paper is optimal offer generation based on cases and game theory. The novel similarity assessment method is proposed and failed cases are particularly mentioned. Simulation of prototype system demonstrates the effectiveness of the method.